diff options
Diffstat (limited to 'fs')
222 files changed, 4610 insertions, 3512 deletions
diff --git a/fs/9p/vfs_addr.c b/fs/9p/vfs_addr.c index c72e9f8f5f32..9a10e68c5f30 100644 --- a/fs/9p/vfs_addr.c +++ b/fs/9p/vfs_addr.c @@ -43,6 +43,11 @@ static void v9fs_req_issue_op(struct netfs_read_subrequest *subreq) iov_iter_xarray(&to, READ, &rreq->mapping->i_pages, pos, len); total = p9_client_read(fid, pos, &to, &err); + + /* if we just extended the file size, any portion not in + * cache won't be on server and is zeroes */ + __set_bit(NETFS_SREQ_CLEAR_TAIL, &subreq->flags); + netfs_subreq_terminated(subreq, err ?: total, false); } diff --git a/fs/9p/vfs_file.c b/fs/9p/vfs_file.c index be72ad9edb3e..2573c08f335c 100644 --- a/fs/9p/vfs_file.c +++ b/fs/9p/vfs_file.c @@ -115,7 +115,6 @@ out_error: static int v9fs_file_lock(struct file *filp, int cmd, struct file_lock *fl) { - int res = 0; struct inode *inode = file_inode(filp); p9_debug(P9_DEBUG_VFS, "filp: %p lock: %p\n", filp, fl); @@ -125,7 +124,7 @@ static int v9fs_file_lock(struct file *filp, int cmd, struct file_lock *fl) invalidate_mapping_pages(&inode->i_data, 0, -1); } - return res; + return 0; } static int v9fs_file_do_lock(struct file *filp, int cmd, struct file_lock *fl) @@ -140,8 +139,7 @@ static int v9fs_file_do_lock(struct file *filp, int cmd, struct file_lock *fl) fid = filp->private_data; BUG_ON(fid == NULL); - if ((fl->fl_flags & FL_POSIX) != FL_POSIX) - BUG(); + BUG_ON((fl->fl_flags & FL_POSIX) != FL_POSIX); res = locks_lock_file_wait(filp, fl); if (res < 0) diff --git a/fs/9p/vfs_inode_dotl.c b/fs/9p/vfs_inode_dotl.c index cae301d09cd3..d17502a738a9 100644 --- a/fs/9p/vfs_inode_dotl.c +++ b/fs/9p/vfs_inode_dotl.c @@ -552,7 +552,10 @@ int v9fs_vfs_setattr_dotl(struct user_namespace *mnt_userns, { int retval, use_dentry = 0; struct p9_fid *fid = NULL; - struct p9_iattr_dotl p9attr; + struct p9_iattr_dotl p9attr = { + .uid = INVALID_UID, + .gid = INVALID_GID, + }; struct inode *inode = d_inode(dentry); p9_debug(P9_DEBUG_VFS, "\n"); @@ -562,14 +565,22 @@ int v9fs_vfs_setattr_dotl(struct user_namespace *mnt_userns, return retval; p9attr.valid = v9fs_mapped_iattr_valid(iattr->ia_valid); - p9attr.mode = iattr->ia_mode; - p9attr.uid = iattr->ia_uid; - p9attr.gid = iattr->ia_gid; - p9attr.size = iattr->ia_size; - p9attr.atime_sec = iattr->ia_atime.tv_sec; - p9attr.atime_nsec = iattr->ia_atime.tv_nsec; - p9attr.mtime_sec = iattr->ia_mtime.tv_sec; - p9attr.mtime_nsec = iattr->ia_mtime.tv_nsec; + if (iattr->ia_valid & ATTR_MODE) + p9attr.mode = iattr->ia_mode; + if (iattr->ia_valid & ATTR_UID) + p9attr.uid = iattr->ia_uid; + if (iattr->ia_valid & ATTR_GID) + p9attr.gid = iattr->ia_gid; + if (iattr->ia_valid & ATTR_SIZE) + p9attr.size = iattr->ia_size; + if (iattr->ia_valid & ATTR_ATIME_SET) { + p9attr.atime_sec = iattr->ia_atime.tv_sec; + p9attr.atime_nsec = iattr->ia_atime.tv_nsec; + } + if (iattr->ia_valid & ATTR_MTIME_SET) { + p9attr.mtime_sec = iattr->ia_mtime.tv_sec; + p9attr.mtime_nsec = iattr->ia_mtime.tv_nsec; + } if (iattr->ia_valid & ATTR_FILE) { fid = iattr->ia_file->private_data; diff --git a/fs/Makefile b/fs/Makefile index 84c5e4cdfee5..dab324aea08f 100644 --- a/fs/Makefile +++ b/fs/Makefile @@ -6,6 +6,8 @@ # Rewritten to use lists instead of if-statements. # +obj-$(CONFIG_SYSCTL) += sysctls.o + obj-y := open.o read_write.o file_table.o super.o \ char_dev.o stat.o exec.o pipe.o namei.o fcntl.o \ ioctl.o readdir.o select.o dcache.o inode.o \ diff --git a/fs/adfs/inode.c b/fs/adfs/inode.c index adbb3a1edcbf..5156821bfe6a 100644 --- a/fs/adfs/inode.c +++ b/fs/adfs/inode.c @@ -355,7 +355,6 @@ int adfs_write_inode(struct inode *inode, struct writeback_control *wbc) { struct super_block *sb = inode->i_sb; struct object_info obj; - int ret; obj.indaddr = ADFS_I(inode)->indaddr; obj.name_len = 0; @@ -365,6 +364,5 @@ int adfs_write_inode(struct inode *inode, struct writeback_control *wbc) obj.attr = ADFS_I(inode)->attr; obj.size = inode->i_size; - ret = adfs_dir_update(sb, &obj, wbc->sync_mode == WB_SYNC_ALL); - return ret; + return adfs_dir_update(sb, &obj, wbc->sync_mode == WB_SYNC_ALL); } diff --git a/fs/afs/proc.c b/fs/afs/proc.c index 065a28bfa3f1..e1b863449296 100644 --- a/fs/afs/proc.c +++ b/fs/afs/proc.c @@ -227,7 +227,7 @@ static int afs_proc_cell_volumes_show(struct seq_file *m, void *v) static void *afs_proc_cell_volumes_start(struct seq_file *m, loff_t *_pos) __acquires(cell->proc_lock) { - struct afs_cell *cell = PDE_DATA(file_inode(m->file)); + struct afs_cell *cell = pde_data(file_inode(m->file)); rcu_read_lock(); return seq_hlist_start_head_rcu(&cell->proc_volumes, *_pos); @@ -236,7 +236,7 @@ static void *afs_proc_cell_volumes_start(struct seq_file *m, loff_t *_pos) static void *afs_proc_cell_volumes_next(struct seq_file *m, void *v, loff_t *_pos) { - struct afs_cell *cell = PDE_DATA(file_inode(m->file)); + struct afs_cell *cell = pde_data(file_inode(m->file)); return seq_hlist_next_rcu(v, &cell->proc_volumes, _pos); } @@ -322,7 +322,7 @@ static void *afs_proc_cell_vlservers_start(struct seq_file *m, loff_t *_pos) { struct afs_vl_seq_net_private *priv = m->private; struct afs_vlserver_list *vllist; - struct afs_cell *cell = PDE_DATA(file_inode(m->file)); + struct afs_cell *cell = pde_data(file_inode(m->file)); loff_t pos = *_pos; rcu_read_lock(); @@ -220,9 +220,35 @@ struct aio_kiocb { /*------ sysctl variables----*/ static DEFINE_SPINLOCK(aio_nr_lock); -unsigned long aio_nr; /* current system wide number of aio requests */ -unsigned long aio_max_nr = 0x10000; /* system wide maximum number of aio requests */ +static unsigned long aio_nr; /* current system wide number of aio requests */ +static unsigned long aio_max_nr = 0x10000; /* system wide maximum number of aio requests */ /*----end sysctl variables---*/ +#ifdef CONFIG_SYSCTL +static struct ctl_table aio_sysctls[] = { + { + .procname = "aio-nr", + .data = &aio_nr, + .maxlen = sizeof(aio_nr), + .mode = 0444, + .proc_handler = proc_doulongvec_minmax, + }, + { + .procname = "aio-max-nr", + .data = &aio_max_nr, + .maxlen = sizeof(aio_max_nr), + .mode = 0644, + .proc_handler = proc_doulongvec_minmax, + }, + {} +}; + +static void __init aio_sysctl_init(void) +{ + register_sysctl_init("fs", aio_sysctls); +} +#else +#define aio_sysctl_init() do { } while (0) +#endif static struct kmem_cache *kiocb_cachep; static struct kmem_cache *kioctx_cachep; @@ -275,6 +301,7 @@ static int __init aio_setup(void) kiocb_cachep = KMEM_CACHE(aio_kiocb, SLAB_HWCACHE_ALIGN|SLAB_PANIC); kioctx_cachep = KMEM_CACHE(kioctx,SLAB_HWCACHE_ALIGN|SLAB_PANIC); + aio_sysctl_init(); return 0; } __initcall(aio_setup); diff --git a/fs/binfmt_elf.c b/fs/binfmt_elf.c index f8c7f26f1fbb..605017eb9349 100644 --- a/fs/binfmt_elf.c +++ b/fs/binfmt_elf.c @@ -1116,11 +1116,11 @@ out_free_interp: * independently randomized mmap region (0 load_bias * without MAP_FIXED nor MAP_FIXED_NOREPLACE). */ - if (interpreter) { + alignment = maximum_alignment(elf_phdata, elf_ex->e_phnum); + if (alignment > ELF_MIN_ALIGN) { load_bias = ELF_ET_DYN_BASE; if (current->flags & PF_RANDOMIZE) load_bias += arch_mmap_rnd(); - alignment = maximum_alignment(elf_phdata, elf_ex->e_phnum); if (alignment) load_bias &= ~(alignment - 1); elf_flags |= MAP_FIXED_NOREPLACE; @@ -1585,7 +1585,7 @@ static int fill_psinfo(struct elf_prpsinfo *psinfo, struct task_struct *p, SET_UID(psinfo->pr_uid, from_kuid_munged(cred->user_ns, cred->uid)); SET_GID(psinfo->pr_gid, from_kgid_munged(cred->user_ns, cred->gid)); rcu_read_unlock(); - strncpy(psinfo->pr_fname, p->comm, sizeof(psinfo->pr_fname)); + get_task_comm(psinfo->pr_fname, p); return 0; } diff --git a/fs/binfmt_misc.c b/fs/binfmt_misc.c index e1eae7ea823a..ddea6acbddde 100644 --- a/fs/binfmt_misc.c +++ b/fs/binfmt_misc.c @@ -822,7 +822,11 @@ static int __init init_misc_binfmt(void) int err = register_filesystem(&bm_fs_type); if (!err) insert_binfmt(&misc_format); - return err; + if (!register_sysctl_mount_point("fs/binfmt_misc")) { + pr_warn("Failed to create fs/binfmt_misc sysctl mount point"); + return -ENOMEM; + } + return 0; } static void __exit exit_misc_binfmt(void) diff --git a/fs/btrfs/Kconfig b/fs/btrfs/Kconfig index 520a0f6a7d9e..183e5c4aed34 100644 --- a/fs/btrfs/Kconfig +++ b/fs/btrfs/Kconfig @@ -18,8 +18,7 @@ config BTRFS_FS select RAID6_PQ select XOR_BLOCKS select SRCU - depends on !PPC_256K_PAGES # powerpc - depends on !PAGE_SIZE_256KB # hexagon + depends on PAGE_SIZE_LESS_THAN_256KB help Btrfs is a general purpose copy-on-write filesystem with extents, diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index d6d48ecf823c..409bad3928db 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -12,7 +12,6 @@ #include <linux/writeback.h> #include <linux/pagevec.h> #include <linux/prefetch.h> -#include <linux/cleancache.h> #include <linux/fsverity.h> #include "misc.h" #include "extent_io.h" @@ -3578,15 +3577,6 @@ int btrfs_do_readpage(struct page *page, struct extent_map **em_cached, goto out; } - if (!PageUptodate(page)) { - if (cleancache_get_page(page) == 0) { - BUG_ON(blocksize != PAGE_SIZE); - unlock_extent(tree, start, end); - unlock_page(page); - goto out; - } - } - if (page->index == last_byte >> PAGE_SHIFT) { size_t zero_offset = offset_in_page(last_byte); diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c index 0ec09fe01be6..4d947ba32da9 100644 --- a/fs/btrfs/super.c +++ b/fs/btrfs/super.c @@ -23,7 +23,6 @@ #include <linux/miscdevice.h> #include <linux/magic.h> #include <linux/slab.h> -#include <linux/cleancache.h> #include <linux/ratelimit.h> #include <linux/crc32c.h> #include <linux/btrfs.h> @@ -1374,7 +1373,6 @@ static int btrfs_fill_super(struct super_block *sb, goto fail_close; } - cleancache_init_fs(sb); sb->s_flags |= SB_ACTIVE; return 0; diff --git a/fs/cachefiles/cache.c b/fs/cachefiles/cache.c index ce4d4785003c..7077f72e6f47 100644 --- a/fs/cachefiles/cache.c +++ b/fs/cachefiles/cache.c @@ -49,11 +49,19 @@ int cachefiles_add_cache(struct cachefiles_cache *cache) goto error_unsupported; } - /* check parameters */ + /* Check features of the backing filesystem: + * - Directories must support looking up and directory creation + * - We create tmpfiles to handle invalidation + * - We use xattrs to store metadata + * - We need to be able to query the amount of space available + * - We want to be able to sync the filesystem when stopping the cache + * - We use DIO to/from pages, so the blocksize mustn't be too big. + */ ret = -EOPNOTSUPP; if (d_is_negative(root) || !d_backing_inode(root)->i_op->lookup || !d_backing_inode(root)->i_op->mkdir || + !d_backing_inode(root)->i_op->tmpfile || !(d_backing_inode(root)->i_opflags & IOP_XATTR) || !root->d_sb->s_op->statfs || !root->d_sb->s_op->sync_fs || @@ -84,9 +92,7 @@ int cachefiles_add_cache(struct cachefiles_cache *cache) goto error_unsupported; cache->bsize = stats.f_bsize; - cache->bshift = 0; - if (stats.f_bsize < PAGE_SIZE) - cache->bshift = PAGE_SHIFT - ilog2(stats.f_bsize); + cache->bshift = ilog2(stats.f_bsize); _debug("blksize %u (shift %u)", cache->bsize, cache->bshift); @@ -106,7 +112,6 @@ int cachefiles_add_cache(struct cachefiles_cache *cache) (unsigned long long) cache->fcull, (unsigned long long) cache->fstop); - stats.f_blocks >>= cache->bshift; do_div(stats.f_blocks, 100); cache->bstop = stats.f_blocks * cache->bstop_percent; cache->bcull = stats.f_blocks * cache->bcull_percent; @@ -209,7 +214,7 @@ int cachefiles_has_space(struct cachefiles_cache *cache, return ret; } - b_avail = stats.f_bavail >> cache->bshift; + b_avail = stats.f_bavail; b_writing = atomic_long_read(&cache->b_writing); if (b_avail > b_writing) b_avail -= b_writing; diff --git a/fs/cachefiles/daemon.c b/fs/cachefiles/daemon.c index 40a792421fc1..7ac04ee2c0a0 100644 --- a/fs/cachefiles/daemon.c +++ b/fs/cachefiles/daemon.c @@ -703,6 +703,17 @@ static int cachefiles_daemon_bind(struct cachefiles_cache *cache, char *args) return -EBUSY; } + /* Make sure we have copies of the tag string */ + if (!cache->tag) { + /* + * The tag string is released by the fops->release() + * function, so we don't release it on error here + */ + cache->tag = kstrdup("CacheFiles", GFP_KERNEL); + if (!cache->tag) + return -ENOMEM; + } + return cachefiles_add_cache(cache); } diff --git a/fs/cachefiles/internal.h b/fs/cachefiles/internal.h index 8dd54d9375b6..c793d33b0224 100644 --- a/fs/cachefiles/internal.h +++ b/fs/cachefiles/internal.h @@ -86,7 +86,7 @@ struct cachefiles_cache { unsigned bcull_percent; /* when to start culling (% blocks) */ unsigned bstop_percent; /* when to stop allocating (% blocks) */ unsigned bsize; /* cache's block size */ - unsigned bshift; /* min(ilog2(PAGE_SIZE / bsize), 0) */ + unsigned bshift; /* ilog2(bsize) */ uint64_t frun; /* when to stop culling */ uint64_t fcull; /* when to start culling */ uint64_t fstop; /* when to stop allocating */ diff --git a/fs/cachefiles/io.c b/fs/cachefiles/io.c index 60b1eac2ce78..04eb52736990 100644 --- a/fs/cachefiles/io.c +++ b/fs/cachefiles/io.c @@ -264,7 +264,7 @@ static int cachefiles_write(struct netfs_cache_resources *cres, ki->term_func = term_func; ki->term_func_priv = term_func_priv; ki->was_async = true; - ki->b_writing = (len + (1 << cache->bshift)) >> cache->bshift; + ki->b_writing = (len + (1 << cache->bshift) - 1) >> cache->bshift; if (ki->term_func) ki->iocb.ki_complete = cachefiles_write_complete; diff --git a/fs/cachefiles/namei.c b/fs/cachefiles/namei.c index 9bd692870617..f256c8aff7bb 100644 --- a/fs/cachefiles/namei.c +++ b/fs/cachefiles/namei.c @@ -25,7 +25,9 @@ static bool __cachefiles_mark_inode_in_use(struct cachefiles_object *object, trace_cachefiles_mark_active(object, inode); can_use = true; } else { - pr_notice("cachefiles: Inode already in use: %pd\n", dentry); + trace_cachefiles_mark_failed(object, inode); + pr_notice("cachefiles: Inode already in use: %pd (B=%lx)\n", + dentry, inode->i_ino); } return can_use; @@ -101,6 +103,7 @@ retry: subdir = lookup_one_len(dirname, dir, strlen(dirname)); else subdir = ERR_PTR(ret); + trace_cachefiles_lookup(NULL, dir, subdir); if (IS_ERR(subdir)) { trace_cachefiles_vfs_error(NULL, d_backing_inode(dir), PTR_ERR(subdir), @@ -135,6 +138,7 @@ retry: cachefiles_trace_mkdir_error); goto mkdir_error; } + trace_cachefiles_mkdir(dir, subdir); if (unlikely(d_unhashed(subdir))) { cachefiles_put_directory(subdir); @@ -233,7 +237,7 @@ static int cachefiles_unlink(struct cachefiles_cache *cache, }; int ret; - trace_cachefiles_unlink(object, dentry, why); + trace_cachefiles_unlink(object, d_inode(dentry)->i_ino, why); ret = security_path_unlink(&path, dentry); if (ret < 0) { cachefiles_io_error(cache, "Unlink security error"); @@ -386,7 +390,7 @@ try_again: .new_dir = d_inode(cache->graveyard), .new_dentry = grave, }; - trace_cachefiles_rename(object, rep, grave, why); + trace_cachefiles_rename(object, d_inode(rep)->i_ino, why); ret = cachefiles_inject_read_error(); if (ret == 0) ret = vfs_rename(&rd); @@ -617,7 +621,7 @@ bool cachefiles_look_up_object(struct cachefiles_object *object) object->d_name_len); else dentry = ERR_PTR(ret); - trace_cachefiles_lookup(object, dentry); + trace_cachefiles_lookup(object, fan, dentry); if (IS_ERR(dentry)) { if (dentry == ERR_PTR(-ENOENT)) goto new_file; diff --git a/fs/ceph/addr.c b/fs/ceph/addr.c index b3d9459c9bbd..c98e5238a1b6 100644 --- a/fs/ceph/addr.c +++ b/fs/ceph/addr.c @@ -297,10 +297,6 @@ out: dout("%s: result %d\n", __func__, err); } -static void ceph_init_rreq(struct netfs_read_request *rreq, struct file *file) -{ -} - static void ceph_readahead_cleanup(struct address_space *mapping, void *priv) { struct inode *inode = mapping->host; @@ -312,7 +308,6 @@ static void ceph_readahead_cleanup(struct address_space *mapping, void *priv) } static const struct netfs_read_request_ops ceph_netfs_read_ops = { - .init_rreq = ceph_init_rreq, .is_cache_enabled = ceph_is_cache_enabled, .begin_cache_operation = ceph_begin_cache_operation, .issue_op = ceph_netfs_issue_op, diff --git a/fs/ceph/caps.c b/fs/ceph/caps.c index 7d2c33cdbac6..7d305b974824 100644 --- a/fs/ceph/caps.c +++ b/fs/ceph/caps.c @@ -3376,8 +3376,7 @@ 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; } diff --git a/fs/ceph/file.c b/fs/ceph/file.c index 9d9304e712d9..5b9104b8e453 100644 --- a/fs/ceph/file.c +++ b/fs/ceph/file.c @@ -204,6 +204,8 @@ 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; dout("%s %p %p 0%o (%s)\n", __func__, inode, file, @@ -225,6 +227,9 @@ 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; } @@ -1541,7 +1546,7 @@ 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: @@ -1556,13 +1561,14 @@ again: 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); @@ -1696,7 +1702,7 @@ 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; @@ -1771,10 +1777,10 @@ retry_snap: 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) diff --git a/fs/ceph/metric.c b/fs/ceph/metric.c index c57699d8408d..0fcba68f9a99 100644 --- a/fs/ceph/metric.c +++ b/fs/ceph/metric.c @@ -160,8 +160,6 @@ 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; 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/super.c b/fs/ceph/super.c index bea89bdb534a..bf79f369aec6 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); @@ -146,6 +148,7 @@ enum { Opt_mds_namespace, Opt_recover_session, Opt_source, + Opt_mon_addr, /* string args above */ Opt_dirstat, Opt_rbytes, @@ -159,6 +162,7 @@ enum { Opt_quotadf, Opt_copyfrom, Opt_wsync, + Opt_pagecache, }; enum ceph_recover_session_mode { @@ -197,8 +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), {} }; @@ -228,9 +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. + * + * 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) * - * The source will look like: + * Old device syntax is: * <server_spec>[,<server_spec>...]:[<path>] * where * <server_spec> is <ip>[:<port>] @@ -263,24 +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) { @@ -306,6 +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; @@ -323,6 +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) @@ -455,6 +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(); } @@ -474,6 +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); } @@ -517,6 +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); } @@ -572,15 +696,22 @@ 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, ",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); if (fsopt->rsize != CEPH_MAX_READ_SIZE) @@ -1052,6 +1183,7 @@ 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; @@ -1063,6 +1195,8 @@ 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); @@ -1148,6 +1282,13 @@ 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; } @@ -1325,6 +1466,14 @@ 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 d0142cc5c41b..67f145e1ae7a 100644 --- a/fs/ceph/super.h +++ b/fs/ceph/super.h @@ -24,13 +24,11 @@ #include <linux/fscache.h> #endif -/* f_type in struct statfs */ -#define CEPH_SUPER_MAGIC 0x00c36400 - /* 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 */ @@ -44,6 +42,7 @@ #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 | \ @@ -88,6 +87,8 @@ 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 +98,7 @@ 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 { @@ -534,19 +536,23 @@ 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, diff --git a/fs/cifs/Kconfig b/fs/cifs/Kconfig index 346ae8716deb..3b7e3b9e4fd2 100644 --- a/fs/cifs/Kconfig +++ b/fs/cifs/Kconfig @@ -188,7 +188,7 @@ config CIFS_SMB_DIRECT config CIFS_FSCACHE bool "Provide CIFS client caching support" - depends on CIFS=m && FSCACHE_OLD_API || CIFS=y && FSCACHE_OLD_API=y + depends on CIFS=m && FSCACHE || CIFS=y && FSCACHE=y help Makes CIFS FS-Cache capable. Say Y here if you want your CIFS data to be cached locally on disk through the general filesystem cache diff --git a/fs/cifs/Makefile b/fs/cifs/Makefile index 87fcacdf3de7..cc8fdcb35b71 100644 --- a/fs/cifs/Makefile +++ b/fs/cifs/Makefile @@ -25,7 +25,7 @@ cifs-$(CONFIG_CIFS_DFS_UPCALL) += cifs_dfs_ref.o dfs_cache.o cifs-$(CONFIG_CIFS_SWN_UPCALL) += netlink.o cifs_swn.o -cifs-$(CONFIG_CIFS_FSCACHE) += fscache.o cache.o +cifs-$(CONFIG_CIFS_FSCACHE) += fscache.o cifs-$(CONFIG_CIFS_SMB_DIRECT) += smbdirect.o diff --git a/fs/cifs/cache.c b/fs/cifs/cache.c deleted file mode 100644 index 8be57aaedab6..000000000000 --- a/fs/cifs/cache.c +++ /dev/null @@ -1,105 +0,0 @@ -// SPDX-License-Identifier: LGPL-2.1 -/* - * CIFS filesystem cache index structure definitions - * - * Copyright (c) 2010 Novell, Inc. - * Authors(s): Suresh Jayaraman (sjayaraman@suse.de> - * - */ -#include "fscache.h" -#include "cifs_debug.h" - -/* - * CIFS filesystem definition for FS-Cache - */ -struct fscache_netfs cifs_fscache_netfs = { - .name = "cifs", - .version = 0, -}; - -/* - * Register CIFS for caching with FS-Cache - */ -int cifs_fscache_register(void) -{ - return fscache_register_netfs(&cifs_fscache_netfs); -} - -/* - * Unregister CIFS for caching - */ -void cifs_fscache_unregister(void) -{ - fscache_unregister_netfs(&cifs_fscache_netfs); -} - -/* - * Server object for FS-Cache - */ -const struct fscache_cookie_def cifs_fscache_server_index_def = { - .name = "CIFS.server", - .type = FSCACHE_COOKIE_TYPE_INDEX, -}; - -static enum -fscache_checkaux cifs_fscache_super_check_aux(void *cookie_netfs_data, - const void *data, - uint16_t datalen, - loff_t object_size) -{ - struct cifs_fscache_super_auxdata auxdata; - const struct cifs_tcon *tcon = cookie_netfs_data; - - if (datalen != sizeof(auxdata)) - return FSCACHE_CHECKAUX_OBSOLETE; - - memset(&auxdata, 0, sizeof(auxdata)); - auxdata.resource_id = tcon->resource_id; - auxdata.vol_create_time = tcon->vol_create_time; - auxdata.vol_serial_number = tcon->vol_serial_number; - - if (memcmp(data, &auxdata, datalen) != 0) - return FSCACHE_CHECKAUX_OBSOLETE; - - return FSCACHE_CHECKAUX_OKAY; -} - -/* - * Superblock object for FS-Cache - */ -const struct fscache_cookie_def cifs_fscache_super_index_def = { - .name = "CIFS.super", - .type = FSCACHE_COOKIE_TYPE_INDEX, - .check_aux = cifs_fscache_super_check_aux, -}; - -static enum -fscache_checkaux cifs_fscache_inode_check_aux(void *cookie_netfs_data, - const void *data, - uint16_t datalen, - loff_t object_size) -{ - struct cifs_fscache_inode_auxdata auxdata; - struct cifsInodeInfo *cifsi = cookie_netfs_data; - - if (datalen != sizeof(auxdata)) - return FSCACHE_CHECKAUX_OBSOLETE; - - memset(&auxdata, 0, sizeof(auxdata)); - auxdata.eof = cifsi->server_eof; - auxdata.last_write_time_sec = cifsi->vfs_inode.i_mtime.tv_sec; - auxdata.last_change_time_sec = cifsi->vfs_inode.i_ctime.tv_sec; - auxdata.last_write_time_nsec = cifsi->vfs_inode.i_mtime.tv_nsec; - auxdata.last_change_time_nsec = cifsi->vfs_inode.i_ctime.tv_nsec; - - if (memcmp(data, &auxdata, datalen) != 0) - return FSCACHE_CHECKAUX_OBSOLETE; - - return FSCACHE_CHECKAUX_OKAY; -} - -const struct fscache_cookie_def cifs_fscache_inode_object_def = { - .name = "CIFS.uniqueid", - .type = FSCACHE_COOKIE_TYPE_DATAFILE, - .check_aux = cifs_fscache_inode_check_aux, -}; diff --git a/fs/cifs/cifs_debug.c b/fs/cifs/cifs_debug.c index d282caf9f037..ea00e1a91250 100644 --- a/fs/cifs/cifs_debug.c +++ b/fs/cifs/cifs_debug.c @@ -416,11 +416,17 @@ skip_rdma: from_kuid(&init_user_ns, ses->cred_uid)); spin_lock(&ses->chan_lock); + if (CIFS_CHAN_NEEDS_RECONNECT(ses, 0)) + seq_puts(m, "\tPrimary channel: DISCONNECTED "); + if (ses->chan_count > 1) { seq_printf(m, "\n\n\tExtra Channels: %zu ", ses->chan_count-1); - for (j = 1; j < ses->chan_count; j++) + for (j = 1; j < ses->chan_count; j++) { cifs_dump_channel(m, j, &ses->chans[j]); + if (CIFS_CHAN_NEEDS_RECONNECT(ses, j)) + seq_puts(m, "\tDISCONNECTED "); + } } spin_unlock(&ses->chan_lock); diff --git a/fs/cifs/cifs_spnego.c b/fs/cifs/cifs_spnego.c index 353bd0dd7026..342717bf1dc2 100644 --- a/fs/cifs/cifs_spnego.c +++ b/fs/cifs/cifs_spnego.c @@ -84,9 +84,9 @@ struct key_type cifs_spnego_key_type = { /* get a key struct with a SPNEGO security blob, suitable for session setup */ struct key * -cifs_get_spnego_key(struct cifs_ses *sesInfo) +cifs_get_spnego_key(struct cifs_ses *sesInfo, + struct TCP_Server_Info *server) { - struct TCP_Server_Info *server = cifs_ses_server(sesInfo); struct sockaddr_in *sa = (struct sockaddr_in *) &server->dstaddr; struct sockaddr_in6 *sa6 = (struct sockaddr_in6 *) &server->dstaddr; char *description, *dp; diff --git a/fs/cifs/cifs_spnego.h b/fs/cifs/cifs_spnego.h index e6a0451877d4..7f102ffeb675 100644 --- a/fs/cifs/cifs_spnego.h +++ b/fs/cifs/cifs_spnego.h @@ -29,7 +29,8 @@ struct cifs_spnego_msg { #ifdef __KERNEL__ extern struct key_type cifs_spnego_key_type; -extern struct key *cifs_get_spnego_key(struct cifs_ses *sesInfo); +extern struct key *cifs_get_spnego_key(struct cifs_ses *sesInfo, + struct TCP_Server_Info *server); #endif /* KERNEL */ #endif /* _CIFS_SPNEGO_H */ diff --git a/fs/cifs/cifs_swn.c b/fs/cifs/cifs_swn.c index 23a1ed2fb769..463ebe34892b 100644 --- a/fs/cifs/cifs_swn.c +++ b/fs/cifs/cifs_swn.c @@ -396,11 +396,11 @@ static int cifs_swn_resource_state_changed(struct cifs_swn_reg *swnreg, const ch switch (state) { case CIFS_SWN_RESOURCE_STATE_UNAVAILABLE: cifs_dbg(FYI, "%s: resource name '%s' become unavailable\n", __func__, name); - cifs_ses_mark_for_reconnect(swnreg->tcon->ses); + cifs_reconnect(swnreg->tcon->ses->server, true); break; case CIFS_SWN_RESOURCE_STATE_AVAILABLE: cifs_dbg(FYI, "%s: resource name '%s' become available\n", __func__, name); - cifs_ses_mark_for_reconnect(swnreg->tcon->ses); + cifs_reconnect(swnreg->tcon->ses->server, true); break; case CIFS_SWN_RESOURCE_STATE_UNKNOWN: cifs_dbg(FYI, "%s: resource name '%s' changed to unknown state\n", __func__, name); @@ -498,10 +498,7 @@ static int cifs_swn_reconnect(struct cifs_tcon *tcon, struct sockaddr_storage *a goto unlock; } - spin_lock(&GlobalMid_Lock); - if (tcon->ses->server->tcpStatus != CifsExiting) - tcon->ses->server->tcpStatus = CifsNeedReconnect; - spin_unlock(&GlobalMid_Lock); + cifs_reconnect(tcon->ses->server, false); unlock: mutex_unlock(&tcon->ses->server->srv_mutex); diff --git a/fs/cifs/cifsencrypt.c b/fs/cifs/cifsencrypt.c index d118282071b3..0912d8bbbac1 100644 --- a/fs/cifs/cifsencrypt.c +++ b/fs/cifs/cifsencrypt.c @@ -141,9 +141,13 @@ int cifs_sign_rqst(struct smb_rqst *rqst, struct TCP_Server_Info *server, if ((cifs_pdu == NULL) || (server == NULL)) return -EINVAL; + spin_lock(&cifs_tcp_ses_lock); if (!(cifs_pdu->Flags2 & SMBFLG2_SECURITY_SIGNATURE) || - server->tcpStatus == CifsNeedNegotiate) + server->tcpStatus == CifsNeedNegotiate) { + spin_unlock(&cifs_tcp_ses_lock); return rc; + } + spin_unlock(&cifs_tcp_ses_lock); if (!server->session_estab) { memcpy(cifs_pdu->Signature.SecuritySignature, "BSRSPYL", 8); diff --git a/fs/cifs/cifsfs.c b/fs/cifs/cifsfs.c index dca42aa87d30..199edac0cb59 100644 --- a/fs/cifs/cifsfs.c +++ b/fs/cifs/cifsfs.c @@ -26,6 +26,7 @@ #include <linux/random.h> #include <linux/uuid.h> #include <linux/xattr.h> +#include <uapi/linux/magic.h> #include <net/ipv6.h> #include "cifsfs.h" #include "cifspdu.h" @@ -202,7 +203,7 @@ cifs_read_super(struct super_block *sb) sb->s_time_max = ts.tv_sec; } - sb->s_magic = CIFS_MAGIC_NUMBER; + sb->s_magic = CIFS_SUPER_MAGIC; sb->s_op = &cifs_super_ops; sb->s_xattr = cifs_xattr_handlers; rc = super_setup_bdi(sb); @@ -396,6 +397,9 @@ static void cifs_evict_inode(struct inode *inode) { truncate_inode_pages_final(&inode->i_data); + if (inode->i_state & I_PINNING_FSCACHE_WB) + cifs_fscache_unuse_inode_cookie(inode, true); + cifs_fscache_release_inode_cookie(inode); clear_inode(inode); } @@ -720,6 +724,12 @@ static int cifs_show_stats(struct seq_file *s, struct dentry *root) } #endif +static int cifs_write_inode(struct inode *inode, struct writeback_control *wbc) +{ + fscache_unpin_writeback(wbc, cifs_inode_cookie(inode)); + return 0; +} + static int cifs_drop_inode(struct inode *inode) { struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); @@ -732,6 +742,7 @@ static int cifs_drop_inode(struct inode *inode) static const struct super_operations cifs_super_ops = { .statfs = cifs_statfs, .alloc_inode = cifs_alloc_inode, + .write_inode = cifs_write_inode, .free_inode = cifs_free_inode, .drop_inode = cifs_drop_inode, .evict_inode = cifs_evict_inode, @@ -773,7 +784,7 @@ cifs_get_root(struct smb3_fs_context *ctx, struct super_block *sb) sep = CIFS_DIR_SEP(cifs_sb); dentry = dget(sb->s_root); - p = s = full_path; + s = full_path; do { struct inode *dir = d_inode(dentry); @@ -1624,13 +1635,9 @@ init_cifs(void) goto out_destroy_cifsoplockd_wq; } - rc = cifs_fscache_register(); - if (rc) - goto out_destroy_deferredclose_wq; - rc = cifs_init_inodecache(); if (rc) - goto out_unreg_fscache; + goto out_destroy_deferredclose_wq; rc = cifs_init_mids(); if (rc) @@ -1692,8 +1699,6 @@ out_destroy_mids: cifs_destroy_mids(); out_destroy_inodecache: cifs_destroy_inodecache(); -out_unreg_fscache: - cifs_fscache_unregister(); out_destroy_deferredclose_wq: destroy_workqueue(deferredclose_wq); out_destroy_cifsoplockd_wq: @@ -1729,7 +1734,6 @@ exit_cifs(void) cifs_destroy_request_bufs(); cifs_destroy_mids(); cifs_destroy_inodecache(); - cifs_fscache_unregister(); destroy_workqueue(deferredclose_wq); destroy_workqueue(cifsoplockd_wq); destroy_workqueue(decrypt_wq); diff --git a/fs/cifs/cifsfs.h b/fs/cifs/cifsfs.h index 9e5d9e192ef0..15a5c5db038b 100644 --- a/fs/cifs/cifsfs.h +++ b/fs/cifs/cifsfs.h @@ -152,5 +152,6 @@ extern struct dentry *cifs_smb3_do_mount(struct file_system_type *fs_type, extern const struct export_operations cifs_export_ops; #endif /* CONFIG_CIFS_NFSD_EXPORT */ -#define CIFS_VERSION "2.34" +#define SMB3_PRODUCT_BUILD 35 +#define CIFS_VERSION "2.35" #endif /* _CIFSFS_H */ diff --git a/fs/cifs/cifsglob.h b/fs/cifs/cifsglob.h index be74606724c7..48b343d03430 100644 --- a/fs/cifs/cifsglob.h +++ b/fs/cifs/cifsglob.h @@ -24,8 +24,6 @@ #include "../smbfs_common/smb2pdu.h" #include "smb2pdu.h" -#define CIFS_MAGIC_NUMBER 0xFF534D42 /* the first four bytes of SMB PDUs */ - #define SMB_PATH_MAX 260 #define CIFS_PORT 445 #define RFC1001_PORT 139 @@ -113,7 +111,14 @@ enum statusEnum { CifsGood, CifsExiting, CifsNeedReconnect, - CifsNeedNegotiate + CifsNeedNegotiate, + CifsInNegotiate, + CifsNeedSessSetup, + CifsInSessSetup, + CifsNeedTcon, + CifsInTcon, + CifsNeedFilesInvalidate, + CifsInFilesInvalidate }; enum securityEnum { @@ -263,13 +268,16 @@ struct smb_version_operations { /* check if we need to negotiate */ bool (*need_neg)(struct TCP_Server_Info *); /* negotiate to the server */ - int (*negotiate)(const unsigned int, struct cifs_ses *); + int (*negotiate)(const unsigned int xid, + struct cifs_ses *ses, + struct TCP_Server_Info *server); /* set negotiated write size */ unsigned int (*negotiate_wsize)(struct cifs_tcon *tcon, struct smb3_fs_context *ctx); /* set negotiated read size */ unsigned int (*negotiate_rsize)(struct cifs_tcon *tcon, struct smb3_fs_context *ctx); /* setup smb sessionn */ int (*sess_setup)(const unsigned int, struct cifs_ses *, + struct TCP_Server_Info *server, const struct nls_table *); /* close smb session */ int (*logoff)(const unsigned int, struct cifs_ses *); @@ -414,7 +422,8 @@ struct smb_version_operations { void (*set_lease_key)(struct inode *, struct cifs_fid *); /* generate new lease key */ void (*new_lease_key)(struct cifs_fid *); - int (*generate_signingkey)(struct cifs_ses *); + int (*generate_signingkey)(struct cifs_ses *ses, + struct TCP_Server_Info *server); int (*calc_signature)(struct smb_rqst *, struct TCP_Server_Info *, bool allocate_crypto); int (*set_integrity)(const unsigned int, struct cifs_tcon *tcon, @@ -582,7 +591,7 @@ struct TCP_Server_Info { char server_RFC1001_name[RFC1001_NAME_LEN_WITH_NULL]; struct smb_version_operations *ops; struct smb_version_values *vals; - /* updates to tcpStatus protected by GlobalMid_Lock */ + /* updates to tcpStatus protected by cifs_tcp_ses_lock */ enum statusEnum tcpStatus; /* what we think the status is */ char *hostname; /* hostname portion of UNC string */ struct socket *ssocket; @@ -659,9 +668,6 @@ struct TCP_Server_Info { unsigned int total_read; /* total amount of data read in this pass */ atomic_t in_send; /* requests trying to send */ atomic_t num_waiters; /* blocked waiting to get in sendrecv */ -#ifdef CONFIG_CIFS_FSCACHE - struct fscache_cookie *fscache; /* client index cache cookie */ -#endif #ifdef CONFIG_CIFS_STATS2 atomic_t num_cmds[NUMBER_OF_SMB2_COMMANDS]; /* total requests by cmd */ atomic_t smb2slowcmd[NUMBER_OF_SMB2_COMMANDS]; /* count resps > 1 sec */ @@ -915,12 +921,13 @@ struct cifs_chan { */ struct cifs_ses { struct list_head smb_ses_list; + struct list_head rlist; /* reconnect list */ struct list_head tcon_list; struct cifs_tcon *tcon_ipc; struct mutex session_mutex; struct TCP_Server_Info *server; /* pointer to server info */ int ses_count; /* reference counter */ - enum statusEnum status; /* updates protected by GlobalMid_Lock */ + enum statusEnum status; /* updates protected by cifs_tcp_ses_lock */ unsigned overrideSecFlg; /* if non-zero override global sec flags */ char *serverOS; /* name of operating system underlying server */ char *serverNOS; /* name of network operating system of server */ @@ -939,17 +946,13 @@ struct cifs_ses { struct ntlmssp_auth *ntlmssp; /* ciphertext, flags, server challenge */ enum securityEnum sectype; /* what security flavor was specified? */ bool sign; /* is signing required? */ - bool need_reconnect:1; /* connection reset, uid now invalid */ bool domainAuto:1; - bool binding:1; /* are we binding the session? */ __u16 session_flags; __u8 smb3signingkey[SMB3_SIGN_KEY_SIZE]; __u8 smb3encryptionkey[SMB3_ENC_DEC_KEY_SIZE]; __u8 smb3decryptionkey[SMB3_ENC_DEC_KEY_SIZE]; __u8 preauth_sha_hash[SMB2_PREAUTH_HASH_SIZE]; - __u8 binding_preauth_sha_hash[SMB2_PREAUTH_HASH_SIZE]; - /* * Network interfaces available on the server this session is * connected to. @@ -969,45 +972,34 @@ struct cifs_ses { spinlock_t chan_lock; /* ========= begin: protected by chan_lock ======== */ #define CIFS_MAX_CHANNELS 16 +#define CIFS_ALL_CHANNELS_SET(ses) \ + ((1UL << (ses)->chan_count) - 1) +#define CIFS_ALL_CHANS_NEED_RECONNECT(ses) \ + ((ses)->chans_need_reconnect == CIFS_ALL_CHANNELS_SET(ses)) +#define CIFS_SET_ALL_CHANS_NEED_RECONNECT(ses) \ + ((ses)->chans_need_reconnect = CIFS_ALL_CHANNELS_SET(ses)) +#define CIFS_CHAN_NEEDS_RECONNECT(ses, index) \ + test_bit((index), &(ses)->chans_need_reconnect) + struct cifs_chan chans[CIFS_MAX_CHANNELS]; - struct cifs_chan *binding_chan; size_t chan_count; size_t chan_max; atomic_t chan_seq; /* round robin state */ + + /* + * chans_need_reconnect is a bitmap indicating which of the channels + * under this smb session needs to be reconnected. + * If not multichannel session, only one bit will be used. + * + * We will ask for sess and tcon reconnection only if all the + * channels are marked for needing reconnection. This will + * enable the sessions on top to continue to live till any + * of the channels below are active. + */ + unsigned long chans_need_reconnect; /* ========= end: protected by chan_lock ======== */ }; -/* - * When binding a new channel, we need to access the channel which isn't fully - * established yet. - */ - -static inline -struct cifs_chan *cifs_ses_binding_channel(struct cifs_ses *ses) -{ - if (ses->binding) - return ses->binding_chan; - else - return NULL; -} - -/* - * Returns the server pointer of the session. When binding a new - * channel this returns the last channel which isn't fully established - * yet. - * - * This function should be use for negprot/sess.setup codepaths. For - * the other requests see cifs_pick_channel(). - */ -static inline -struct TCP_Server_Info *cifs_ses_server(struct cifs_ses *ses) -{ - if (ses->binding) - return ses->binding_chan->server; - else - return ses->server; -} - static inline bool cap_unix(struct cifs_ses *ses) { @@ -1117,7 +1109,7 @@ struct cifs_tcon { __u32 max_bytes_copy; #ifdef CONFIG_CIFS_FSCACHE u64 resource_id; /* server resource id */ - struct fscache_cookie *fscache; /* cookie for share */ + struct fscache_volume *fscache; /* cookie for share */ #endif struct list_head pending_opens; /* list of incomplete opens */ struct cached_fid crfid; /* Cached root fid */ diff --git a/fs/cifs/cifspdu.h b/fs/cifs/cifspdu.h index d2ff438fd31f..68b9a436af4b 100644 --- a/fs/cifs/cifspdu.h +++ b/fs/cifs/cifspdu.h @@ -2560,7 +2560,7 @@ typedef struct { __le32 EaSize; /* length of the xattrs */ __u8 ShortNameLength; __u8 Reserved; - __u8 ShortName[12]; + __u8 ShortName[24]; char FileName[1]; } __attribute__((packed)) FILE_BOTH_DIRECTORY_INFO; /* level 0x104 FFrsp data */ diff --git a/fs/cifs/cifsproto.h b/fs/cifs/cifsproto.h index 4f5a3e857df4..d3701295402d 100644 --- a/fs/cifs/cifsproto.h +++ b/fs/cifs/cifsproto.h @@ -131,7 +131,11 @@ extern int SendReceiveBlockingLock(const unsigned int xid, struct smb_hdr *in_buf , struct smb_hdr *out_buf, int *bytes_returned); -extern int cifs_reconnect(struct TCP_Server_Info *server); +void +cifs_mark_tcp_ses_conns_for_reconnect(struct TCP_Server_Info *server, + bool mark_smb_session); +extern int cifs_reconnect(struct TCP_Server_Info *server, + bool mark_smb_session); extern int checkSMB(char *buf, unsigned int len, struct TCP_Server_Info *srvr); extern bool is_valid_oplock_break(char *, struct TCP_Server_Info *); extern bool backup_cred(struct cifs_sb_info *); @@ -164,6 +168,7 @@ extern int small_smb_init_no_tc(const int smb_cmd, const int wct, extern enum securityEnum select_sectype(struct TCP_Server_Info *server, enum securityEnum requested); extern int CIFS_SessSetup(const unsigned int xid, struct cifs_ses *ses, + struct TCP_Server_Info *server, const struct nls_table *nls_cp); extern struct timespec64 cifs_NTtimeToUnix(__le64 utc_nanoseconds_since_1601); extern u64 cifs_UnixTimeToNT(struct timespec64); @@ -293,11 +298,15 @@ extern int cifs_tree_connect(const unsigned int xid, struct cifs_tcon *tcon, const struct nls_table *nlsc); extern int cifs_negotiate_protocol(const unsigned int xid, - struct cifs_ses *ses); + struct cifs_ses *ses, + struct TCP_Server_Info *server); extern int cifs_setup_session(const unsigned int xid, struct cifs_ses *ses, + struct TCP_Server_Info *server, struct nls_table *nls_info); extern int cifs_enable_signing(struct TCP_Server_Info *server, bool mnt_sign_required); -extern int CIFSSMBNegotiate(const unsigned int xid, struct cifs_ses *ses); +extern int CIFSSMBNegotiate(const unsigned int xid, + struct cifs_ses *ses, + struct TCP_Server_Info *server); extern int CIFSTCon(const unsigned int xid, struct cifs_ses *ses, const char *tree, struct cifs_tcon *tcon, @@ -504,8 +513,10 @@ extern int cifs_verify_signature(struct smb_rqst *rqst, extern int setup_ntlmv2_rsp(struct cifs_ses *, const struct nls_table *); extern void cifs_crypto_secmech_release(struct TCP_Server_Info *server); extern int calc_seckey(struct cifs_ses *); -extern int generate_smb30signingkey(struct cifs_ses *); -extern int generate_smb311signingkey(struct cifs_ses *); +extern int generate_smb30signingkey(struct cifs_ses *ses, + struct TCP_Server_Info *server); +extern int generate_smb311signingkey(struct cifs_ses *ses, + struct TCP_Server_Info *server); extern int CIFSSMBCopy(unsigned int xid, struct cifs_tcon *source_tcon, @@ -601,6 +612,19 @@ bool is_server_using_iface(struct TCP_Server_Info *server, bool is_ses_using_iface(struct cifs_ses *ses, struct cifs_server_iface *iface); void cifs_ses_mark_for_reconnect(struct cifs_ses *ses); +unsigned int +cifs_ses_get_chan_index(struct cifs_ses *ses, + struct TCP_Server_Info *server); +void +cifs_chan_set_need_reconnect(struct cifs_ses *ses, + struct TCP_Server_Info *server); +void +cifs_chan_clear_need_reconnect(struct cifs_ses *ses, + struct TCP_Server_Info *server); +bool +cifs_chan_needs_reconnect(struct cifs_ses *ses, + struct TCP_Server_Info *server); + void extract_unc_hostname(const char *unc, const char **h, size_t *len); int copy_path_name(char *dst, const char *src); int smb2_parse_query_directory(struct cifs_tcon *tcon, struct kvec *rsp_iov, @@ -626,6 +650,11 @@ static inline int get_dfs_path(const unsigned int xid, struct cifs_ses *ses, int match_target_ip(struct TCP_Server_Info *server, const char *share, size_t share_len, bool *result); + +int cifs_dfs_query_info_nonascii_quirk(const unsigned int xid, + struct cifs_tcon *tcon, + struct cifs_sb_info *cifs_sb, + const char *dfs_link_path); #endif static inline int cifs_create_options(struct cifs_sb_info *cifs_sb, int options) diff --git a/fs/cifs/cifssmb.c b/fs/cifs/cifssmb.c index 243d17696f06..071e2f21a7db 100644 --- a/fs/cifs/cifssmb.c +++ b/fs/cifs/cifssmb.c @@ -73,6 +73,16 @@ cifs_mark_open_files_invalid(struct cifs_tcon *tcon) struct list_head *tmp; struct list_head *tmp1; + /* only send once per connect */ + spin_lock(&cifs_tcp_ses_lock); + if (tcon->ses->status != CifsGood || + tcon->tidStatus != CifsNeedReconnect) { + spin_unlock(&cifs_tcp_ses_lock); + return; + } + tcon->tidStatus = CifsInFilesInvalidate; + spin_unlock(&cifs_tcp_ses_lock); + /* list all files open on tree connection and mark them invalid */ spin_lock(&tcon->open_file_lock); list_for_each_safe(tmp, tmp1, &tcon->openFileList) { @@ -89,6 +99,11 @@ cifs_mark_open_files_invalid(struct cifs_tcon *tcon) memset(tcon->crfid.fid, 0, sizeof(struct cifs_fid)); mutex_unlock(&tcon->crfid.fid_mutex); + spin_lock(&cifs_tcp_ses_lock); + if (tcon->tidStatus == CifsInFilesInvalidate) + tcon->tidStatus = CifsNeedTcon; + spin_unlock(&cifs_tcp_ses_lock); + /* * BB Add call to invalidate_inodes(sb) for all superblocks mounted * to this tcon. @@ -120,15 +135,18 @@ cifs_reconnect_tcon(struct cifs_tcon *tcon, int smb_command) * only tree disconnect, open, and write, (and ulogoff which does not * have tcon) are allowed as we start force umount */ + spin_lock(&cifs_tcp_ses_lock); if (tcon->tidStatus == CifsExiting) { if (smb_command != SMB_COM_WRITE_ANDX && smb_command != SMB_COM_OPEN_ANDX && smb_command != SMB_COM_TREE_DISCONNECT) { + spin_unlock(&cifs_tcp_ses_lock); cifs_dbg(FYI, "can not send cmd %d while umounting\n", smb_command); return -ENODEV; } } + spin_unlock(&cifs_tcp_ses_lock); retries = server->nr_targets; @@ -148,8 +166,12 @@ cifs_reconnect_tcon(struct cifs_tcon *tcon, int smb_command) } /* are we still trying to reconnect? */ - if (server->tcpStatus != CifsNeedReconnect) + spin_lock(&cifs_tcp_ses_lock); + if (server->tcpStatus != CifsNeedReconnect) { + spin_unlock(&cifs_tcp_ses_lock); break; + } + spin_unlock(&cifs_tcp_ses_lock); if (retries && --retries) continue; @@ -166,31 +188,49 @@ cifs_reconnect_tcon(struct cifs_tcon *tcon, int smb_command) retries = server->nr_targets; } - if (!ses->need_reconnect && !tcon->need_reconnect) + spin_lock(&ses->chan_lock); + if (!cifs_chan_needs_reconnect(ses, server) && !tcon->need_reconnect) { + spin_unlock(&ses->chan_lock); return 0; + } + spin_unlock(&ses->chan_lock); nls_codepage = load_nls_default(); /* - * need to prevent multiple threads trying to simultaneously - * reconnect the same SMB session - */ - mutex_lock(&ses->session_mutex); - - /* * Recheck after acquire mutex. If another thread is negotiating * and the server never sends an answer the socket will be closed * and tcpStatus set to reconnect. */ + spin_lock(&cifs_tcp_ses_lock); if (server->tcpStatus == CifsNeedReconnect) { + spin_unlock(&cifs_tcp_ses_lock); rc = -EHOSTDOWN; - mutex_unlock(&ses->session_mutex); goto out; } + spin_unlock(&cifs_tcp_ses_lock); - rc = cifs_negotiate_protocol(0, ses); - if (rc == 0 && ses->need_reconnect) - rc = cifs_setup_session(0, ses, nls_codepage); + /* + * need to prevent multiple threads trying to simultaneously + * reconnect the same SMB session + */ + spin_lock(&ses->chan_lock); + if (!cifs_chan_needs_reconnect(ses, server)) { + spin_unlock(&ses->chan_lock); + + /* this means that we only need to tree connect */ + if (tcon->need_reconnect) + goto skip_sess_setup; + + rc = -EHOSTDOWN; + goto out; + } + spin_unlock(&ses->chan_lock); + + mutex_lock(&ses->session_mutex); + rc = cifs_negotiate_protocol(0, ses, server); + if (!rc) + rc = cifs_setup_session(0, ses, server, nls_codepage); /* do we need to reconnect tcon? */ if (rc || !tcon->need_reconnect) { @@ -198,6 +238,7 @@ cifs_reconnect_tcon(struct cifs_tcon *tcon, int smb_command) goto out; } +skip_sess_setup: cifs_mark_open_files_invalid(tcon); rc = cifs_tree_connect(0, tcon, nls_codepage); mutex_unlock(&ses->session_mutex); @@ -337,8 +378,13 @@ static int smb_init_no_reconnect(int smb_command, int wct, struct cifs_tcon *tcon, void **request_buf, void **response_buf) { - if (tcon->ses->need_reconnect || tcon->need_reconnect) + spin_lock(&tcon->ses->chan_lock); + if (cifs_chan_needs_reconnect(tcon->ses, tcon->ses->server) || + tcon->need_reconnect) { + spin_unlock(&tcon->ses->chan_lock); return -EHOSTDOWN; + } + spin_unlock(&tcon->ses->chan_lock); return __smb_init(smb_command, wct, tcon, request_buf, response_buf); } @@ -476,14 +522,15 @@ should_set_ext_sec_flag(enum securityEnum sectype) } int -CIFSSMBNegotiate(const unsigned int xid, struct cifs_ses *ses) +CIFSSMBNegotiate(const unsigned int xid, + struct cifs_ses *ses, + struct TCP_Server_Info *server) { NEGOTIATE_REQ *pSMB; NEGOTIATE_RSP *pSMBr; int rc = 0; int bytes_returned; int i; - struct TCP_Server_Info *server = ses->server; u16 count; if (!server) { @@ -600,8 +647,12 @@ CIFSSMBTDis(const unsigned int xid, struct cifs_tcon *tcon) * the tcon is no longer on the list, so no need to take lock before * checking this. */ - if ((tcon->need_reconnect) || (tcon->ses->need_reconnect)) - return 0; + spin_lock(&tcon->ses->chan_lock); + if ((tcon->need_reconnect) || CIFS_ALL_CHANS_NEED_RECONNECT(tcon->ses)) { + spin_unlock(&tcon->ses->chan_lock); + return -EIO; + } + spin_unlock(&tcon->ses->chan_lock); rc = small_smb_init(SMB_COM_TREE_DISCONNECT, 0, tcon, (void **)&smb_buffer); @@ -696,9 +747,14 @@ CIFSSMBLogoff(const unsigned int xid, struct cifs_ses *ses) return -EIO; mutex_lock(&ses->session_mutex); - if (ses->need_reconnect) + spin_lock(&ses->chan_lock); + if (CIFS_ALL_CHANS_NEED_RECONNECT(ses)) { + spin_unlock(&ses->chan_lock); goto session_already_dead; /* no need to send SMBlogoff if uid already closed due to reconnect */ + } + spin_unlock(&ses->chan_lock); + rc = small_smb_init(SMB_COM_LOGOFF_ANDX, 2, NULL, (void **)&pSMB); if (rc) { mutex_unlock(&ses->session_mutex); @@ -1401,7 +1457,7 @@ cifs_readv_receive(struct TCP_Server_Info *server, struct mid_q_entry *mid) if (server->ops->is_session_expired && server->ops->is_session_expired(buf)) { - cifs_reconnect(server); + cifs_reconnect(server, true); return -1; } diff --git a/fs/cifs/connect.c b/fs/cifs/connect.c index 1060164b984a..11a22a30ee14 100644 --- a/fs/cifs/connect.c +++ b/fs/cifs/connect.c @@ -166,36 +166,57 @@ static void cifs_resolve_server(struct work_struct *work) * Mark all sessions and tcons for reconnect. * * @server needs to be previously set to CifsNeedReconnect. + * */ -static void cifs_mark_tcp_ses_conns_for_reconnect(struct TCP_Server_Info *server) +void +cifs_mark_tcp_ses_conns_for_reconnect(struct TCP_Server_Info *server, + bool mark_smb_session) { + struct TCP_Server_Info *pserver; struct cifs_ses *ses; struct cifs_tcon *tcon; struct mid_q_entry *mid, *nmid; struct list_head retry_list; - struct TCP_Server_Info *pserver; server->maxBuf = 0; server->max_read = 0; - cifs_dbg(FYI, "Mark tcp session as need reconnect\n"); - trace_smb3_reconnect(server->CurrentMid, server->conn_id, server->hostname); /* * before reconnecting the tcp session, mark the smb session (uid) and the tid bad so they * are not used until reconnected. */ - cifs_dbg(FYI, "%s: marking sessions and tcons for reconnect\n", __func__); + cifs_dbg(FYI, "%s: marking necessary sessions and tcons for reconnect\n", __func__); /* If server is a channel, select the primary channel */ pserver = CIFS_SERVER_IS_CHAN(server) ? server->primary_server : server; + spin_lock(&cifs_tcp_ses_lock); list_for_each_entry(ses, &pserver->smb_ses_list, smb_ses_list) { - ses->need_reconnect = true; - list_for_each_entry(tcon, &ses->tcon_list, tcon_list) + spin_lock(&ses->chan_lock); + if (!mark_smb_session && cifs_chan_needs_reconnect(ses, server)) + goto next_session; + + if (mark_smb_session) + CIFS_SET_ALL_CHANS_NEED_RECONNECT(ses); + else + cifs_chan_set_need_reconnect(ses, server); + + /* If all channels need reconnect, then tcon needs reconnect */ + if (!mark_smb_session && !CIFS_ALL_CHANS_NEED_RECONNECT(ses)) + goto next_session; + + ses->status = CifsNeedReconnect; + + list_for_each_entry(tcon, &ses->tcon_list, tcon_list) { tcon->need_reconnect = true; + tcon->tidStatus = CifsNeedReconnect; + } if (ses->tcon_ipc) ses->tcon_ipc->need_reconnect = true; + +next_session: + spin_unlock(&ses->chan_lock); } spin_unlock(&cifs_tcp_ses_lock); @@ -248,16 +269,21 @@ static void cifs_mark_tcp_ses_conns_for_reconnect(struct TCP_Server_Info *server static bool cifs_tcp_ses_needs_reconnect(struct TCP_Server_Info *server, int num_targets) { - spin_lock(&GlobalMid_Lock); + spin_lock(&cifs_tcp_ses_lock); server->nr_targets = num_targets; if (server->tcpStatus == CifsExiting) { /* the demux thread will exit normally next time through the loop */ - spin_unlock(&GlobalMid_Lock); + spin_unlock(&cifs_tcp_ses_lock); wake_up(&server->response_q); return false; } + + cifs_dbg(FYI, "Mark tcp session as need reconnect\n"); + trace_smb3_reconnect(server->CurrentMid, server->conn_id, + server->hostname); server->tcpStatus = CifsNeedReconnect; - spin_unlock(&GlobalMid_Lock); + + spin_unlock(&cifs_tcp_ses_lock); return true; } @@ -268,15 +294,21 @@ static bool cifs_tcp_ses_needs_reconnect(struct TCP_Server_Info *server, int num * mark all smb sessions as reconnecting for tcp session * reconnect tcp session * wake up waiters on reconnection? - (not needed currently) + * + * if mark_smb_session is passed as true, unconditionally mark + * the smb session (and tcon) for reconnect as well. This value + * doesn't really matter for non-multichannel scenario. + * */ -static int __cifs_reconnect(struct TCP_Server_Info *server) +static int __cifs_reconnect(struct TCP_Server_Info *server, + bool mark_smb_session) { int rc = 0; if (!cifs_tcp_ses_needs_reconnect(server, 1)) return 0; - cifs_mark_tcp_ses_conns_for_reconnect(server); + cifs_mark_tcp_ses_conns_for_reconnect(server, mark_smb_session); do { try_to_freeze(); @@ -299,17 +331,20 @@ static int __cifs_reconnect(struct TCP_Server_Info *server) } else { atomic_inc(&tcpSesReconnectCount); set_credits(server, 1); - spin_lock(&GlobalMid_Lock); + spin_lock(&cifs_tcp_ses_lock); if (server->tcpStatus != CifsExiting) server->tcpStatus = CifsNeedNegotiate; - spin_unlock(&GlobalMid_Lock); + spin_unlock(&cifs_tcp_ses_lock); cifs_swn_reset_server_dstaddr(server); mutex_unlock(&server->srv_mutex); + mod_delayed_work(cifsiod_wq, &server->reconnect, 0); } } while (server->tcpStatus == CifsNeedReconnect); + spin_lock(&cifs_tcp_ses_lock); if (server->tcpStatus == CifsNeedNegotiate) mod_delayed_work(cifsiod_wq, &server->echo, 0); + spin_unlock(&cifs_tcp_ses_lock); wake_up(&server->response_q); return rc; @@ -371,7 +406,9 @@ static int reconnect_target_unlocked(struct TCP_Server_Info *server, struct dfs_ return rc; } -static int reconnect_dfs_server(struct TCP_Server_Info *server) +static int +reconnect_dfs_server(struct TCP_Server_Info *server, + bool mark_smb_session) { int rc = 0; const char *refpath = server->current_fullpath + 1; @@ -395,7 +432,7 @@ static int reconnect_dfs_server(struct TCP_Server_Info *server) if (!cifs_tcp_ses_needs_reconnect(server, num_targets)) return 0; - cifs_mark_tcp_ses_conns_for_reconnect(server); + cifs_mark_tcp_ses_conns_for_reconnect(server, mark_smb_session); do { try_to_freeze(); @@ -416,12 +453,13 @@ static int reconnect_dfs_server(struct TCP_Server_Info *server) */ atomic_inc(&tcpSesReconnectCount); set_credits(server, 1); - spin_lock(&GlobalMid_Lock); + spin_lock(&cifs_tcp_ses_lock); if (server->tcpStatus != CifsExiting) server->tcpStatus = CifsNeedNegotiate; - spin_unlock(&GlobalMid_Lock); + spin_unlock(&cifs_tcp_ses_lock); cifs_swn_reset_server_dstaddr(server); mutex_unlock(&server->srv_mutex); + mod_delayed_work(cifsiod_wq, &server->reconnect, 0); } while (server->tcpStatus == CifsNeedReconnect); if (target_hint) @@ -430,29 +468,32 @@ static int reconnect_dfs_server(struct TCP_Server_Info *server) dfs_cache_free_tgts(&tl); /* Need to set up echo worker again once connection has been established */ + spin_lock(&cifs_tcp_ses_lock); if (server->tcpStatus == CifsNeedNegotiate) mod_delayed_work(cifsiod_wq, &server->echo, 0); + spin_unlock(&cifs_tcp_ses_lock); + wake_up(&server->response_q); return rc; } -int cifs_reconnect(struct TCP_Server_Info *server) +int cifs_reconnect(struct TCP_Server_Info *server, bool mark_smb_session) { /* If tcp session is not an dfs connection, then reconnect to last target server */ spin_lock(&cifs_tcp_ses_lock); if (!server->is_dfs_conn || !server->origin_fullpath || !server->leaf_fullpath) { spin_unlock(&cifs_tcp_ses_lock); - return __cifs_reconnect(server); + return __cifs_reconnect(server, mark_smb_session); } spin_unlock(&cifs_tcp_ses_lock); - return reconnect_dfs_server(server); + return reconnect_dfs_server(server, mark_smb_session); } #else -int cifs_reconnect(struct TCP_Server_Info *server) +int cifs_reconnect(struct TCP_Server_Info *server, bool mark_smb_session) { - return __cifs_reconnect(server); + return __cifs_reconnect(server, mark_smb_session); } #endif @@ -534,15 +575,18 @@ server_unresponsive(struct TCP_Server_Info *server) * 65s kernel_recvmsg times out, and we see that we haven't gotten * a response in >60s. */ + spin_lock(&cifs_tcp_ses_lock); if ((server->tcpStatus == CifsGood || server->tcpStatus == CifsNeedNegotiate) && (!server->ops->can_echo || server->ops->can_echo(server)) && time_after(jiffies, server->lstrp + 3 * server->echo_interval)) { + spin_unlock(&cifs_tcp_ses_lock); cifs_server_dbg(VFS, "has not responded in %lu seconds. Reconnecting...\n", (3 * server->echo_interval) / HZ); - cifs_reconnect(server); + cifs_reconnect(server, false); return true; } + spin_unlock(&cifs_tcp_ses_lock); return false; } @@ -576,7 +620,7 @@ cifs_readv_from_socket(struct TCP_Server_Info *server, struct msghdr *smb_msg) /* reconnect if no credits and no requests in flight */ if (zero_credits(server)) { - cifs_reconnect(server); + cifs_reconnect(server, false); return -ECONNABORTED; } @@ -587,13 +631,17 @@ cifs_readv_from_socket(struct TCP_Server_Info *server, struct msghdr *smb_msg) else length = sock_recvmsg(server->ssocket, smb_msg, 0); - if (server->tcpStatus == CifsExiting) + spin_lock(&cifs_tcp_ses_lock); + if (server->tcpStatus == CifsExiting) { + spin_unlock(&cifs_tcp_ses_lock); return -ESHUTDOWN; + } if (server->tcpStatus == CifsNeedReconnect) { - cifs_reconnect(server); + spin_unlock(&cifs_tcp_ses_lock); return -ECONNABORTED; } + spin_unlock(&cifs_tcp_ses_lock); if (length == -ERESTARTSYS || length == -EAGAIN || @@ -610,7 +658,7 @@ cifs_readv_from_socket(struct TCP_Server_Info *server, struct msghdr *smb_msg) if (length <= 0) { cifs_dbg(FYI, "Received no data or error: %d\n", length); - cifs_reconnect(server); + cifs_reconnect(server, false); return -ECONNABORTED; } } @@ -689,11 +737,11 @@ is_smb_response(struct TCP_Server_Info *server, unsigned char type) * initialize frame). */ cifs_set_port((struct sockaddr *)&server->dstaddr, CIFS_PORT); - cifs_reconnect(server); + cifs_reconnect(server, true); break; default: cifs_server_dbg(VFS, "RFC 1002 unknown response type 0x%x\n", type); - cifs_reconnect(server); + cifs_reconnect(server, true); } return false; @@ -771,9 +819,9 @@ static void clean_demultiplex_info(struct TCP_Server_Info *server) cancel_delayed_work_sync(&server->echo); cancel_delayed_work_sync(&server->resolve); - spin_lock(&GlobalMid_Lock); + spin_lock(&cifs_tcp_ses_lock); server->tcpStatus = CifsExiting; - spin_unlock(&GlobalMid_Lock); + spin_unlock(&cifs_tcp_ses_lock); wake_up_all(&server->response_q); /* check if we have blocked requests that need to free */ @@ -866,7 +914,7 @@ standard_receive3(struct TCP_Server_Info *server, struct mid_q_entry *mid) if (pdu_length > CIFSMaxBufSize + MAX_HEADER_SIZE(server) - server->vals->header_preamble_size) { cifs_server_dbg(VFS, "SMB response too long (%u bytes)\n", pdu_length); - cifs_reconnect(server); + cifs_reconnect(server, true); return -ECONNABORTED; } @@ -913,7 +961,7 @@ cifs_handle_standard(struct TCP_Server_Info *server, struct mid_q_entry *mid) if (server->ops->is_session_expired && server->ops->is_session_expired(buf)) { - cifs_reconnect(server); + cifs_reconnect(server, true); return -1; } @@ -1017,7 +1065,7 @@ next_pdu: server->vals->header_preamble_size) { cifs_server_dbg(VFS, "SMB response too short (%u bytes)\n", server->pdu_size); - cifs_reconnect(server); + cifs_reconnect(server, true); continue; } @@ -1069,7 +1117,7 @@ next_pdu: server->ops->is_status_io_timeout(buf)) { num_io_timeout++; if (num_io_timeout > NUM_STATUS_IO_TIMEOUT) { - cifs_reconnect(server); + cifs_reconnect(server, false); num_io_timeout = 0; continue; } @@ -1139,7 +1187,7 @@ next_pdu: } memalloc_noreclaim_restore(noreclaim_flag); - module_put_and_exit(0); + module_put_and_kthread_exit(0); } /* @@ -1390,16 +1438,12 @@ cifs_put_tcp_session(struct TCP_Server_Info *server, int from_reconnect) else cancel_delayed_work_sync(&server->reconnect); - spin_lock(&GlobalMid_Lock); + spin_lock(&cifs_tcp_ses_lock); server->tcpStatus = CifsExiting; - spin_unlock(&GlobalMid_Lock); + spin_unlock(&cifs_tcp_ses_lock); cifs_crypto_secmech_release(server); - /* fscache server cookies are based on primary channel only */ - if (!CIFS_SERVER_IS_CHAN(server)) - cifs_fscache_release_client_cookie(server); - kfree(server->session_key.response); server->session_key.response = NULL; server->session_key.len = 0; @@ -1545,7 +1589,9 @@ smbd_connected: * to the struct since the kernel thread not created yet * no need to spinlock this update of tcpStatus */ + spin_lock(&cifs_tcp_ses_lock); tcp_ses->tcpStatus = CifsNeedNegotiate; + spin_unlock(&cifs_tcp_ses_lock); if ((ctx->max_credits < 20) || (ctx->max_credits > 60000)) tcp_ses->max_credits = SMB2_MAX_CREDITS_AVAILABLE; @@ -1559,14 +1605,6 @@ smbd_connected: list_add(&tcp_ses->tcp_ses_list, &cifs_tcp_ses_list); spin_unlock(&cifs_tcp_ses_lock); - /* fscache server cookies are based on primary channel only */ - if (!CIFS_SERVER_IS_CHAN(tcp_ses)) - cifs_fscache_get_client_cookie(tcp_ses); -#ifdef CONFIG_CIFS_FSCACHE - else - tcp_ses->fscache = tcp_ses->primary_server->fscache; -#endif /* CONFIG_CIFS_FSCACHE */ - /* queue echo request delayed work */ queue_delayed_work(cifsiod_wq, &tcp_ses->echo, tcp_ses->echo_interval); @@ -1762,15 +1800,13 @@ void cifs_put_smb_ses(struct cifs_ses *ses) spin_unlock(&cifs_tcp_ses_lock); return; } - spin_unlock(&cifs_tcp_ses_lock); /* ses_count can never go negative */ WARN_ON(ses->ses_count < 0); - spin_lock(&GlobalMid_Lock); if (ses->status == CifsGood) ses->status = CifsExiting; - spin_unlock(&GlobalMid_Lock); + spin_unlock(&cifs_tcp_ses_lock); cifs_free_ipc(ses); @@ -1789,7 +1825,6 @@ void cifs_put_smb_ses(struct cifs_ses *ses) spin_lock(&ses->chan_lock); chan_count = ses->chan_count; - spin_unlock(&ses->chan_lock); /* close any extra channels */ if (chan_count > 1) { @@ -1806,6 +1841,7 @@ void cifs_put_smb_ses(struct cifs_ses *ses) ses->chans[i].server = NULL; } } + spin_unlock(&ses->chan_lock); sesInfoFree(ses); cifs_put_tcp_session(server, 0); @@ -1987,11 +2023,13 @@ cifs_get_smb_ses(struct TCP_Server_Info *server, struct smb3_fs_context *ctx) cifs_dbg(FYI, "Existing smb sess found (status=%d)\n", ses->status); - mutex_lock(&ses->session_mutex); - if (ses->need_reconnect) { + spin_lock(&ses->chan_lock); + if (cifs_chan_needs_reconnect(ses, server)) { + spin_unlock(&ses->chan_lock); cifs_dbg(FYI, "Session needs reconnect\n"); - rc = cifs_negotiate_protocol(xid, ses); + mutex_lock(&ses->session_mutex); + rc = cifs_negotiate_protocol(xid, ses, server); if (rc) { mutex_unlock(&ses->session_mutex); /* problem -- put our ses reference */ @@ -2000,7 +2038,7 @@ cifs_get_smb_ses(struct TCP_Server_Info *server, struct smb3_fs_context *ctx) return ERR_PTR(rc); } - rc = cifs_setup_session(xid, ses, + rc = cifs_setup_session(xid, ses, server, ctx->local_nls); if (rc) { mutex_unlock(&ses->session_mutex); @@ -2009,8 +2047,11 @@ cifs_get_smb_ses(struct TCP_Server_Info *server, struct smb3_fs_context *ctx) free_xid(xid); return ERR_PTR(rc); } + mutex_unlock(&ses->session_mutex); + + spin_lock(&ses->chan_lock); } - mutex_unlock(&ses->session_mutex); + spin_unlock(&ses->chan_lock); /* existing SMB ses has a server reference already */ cifs_put_tcp_session(server, 0); @@ -2060,28 +2101,35 @@ cifs_get_smb_ses(struct TCP_Server_Info *server, struct smb3_fs_context *ctx) ses->sectype = ctx->sectype; ses->sign = ctx->sign; - mutex_lock(&ses->session_mutex); /* add server as first channel */ spin_lock(&ses->chan_lock); ses->chans[0].server = server; ses->chan_count = 1; ses->chan_max = ctx->multichannel ? ctx->max_channels:1; + ses->chans_need_reconnect = 1; spin_unlock(&ses->chan_lock); - rc = cifs_negotiate_protocol(xid, ses); + mutex_lock(&ses->session_mutex); + rc = cifs_negotiate_protocol(xid, ses, server); if (!rc) - rc = cifs_setup_session(xid, ses, ctx->local_nls); + rc = cifs_setup_session(xid, ses, server, ctx->local_nls); + mutex_unlock(&ses->session_mutex); /* each channel uses a different signing key */ + spin_lock(&ses->chan_lock); memcpy(ses->chans[0].signkey, ses->smb3signingkey, sizeof(ses->smb3signingkey)); + spin_unlock(&ses->chan_lock); - mutex_unlock(&ses->session_mutex); if (rc) goto get_ses_fail; - /* success, put it on the list and add it as first channel */ + /* + * success, put it on the list and add it as first channel + * note: the session becomes active soon after this. So you'll + * need to lock before changing something in the session. + */ spin_lock(&cifs_tcp_ses_lock); list_add(&ses->smb_ses_list, &server->smb_ses_list); spin_unlock(&cifs_tcp_ses_lock); @@ -2161,6 +2209,9 @@ cifs_put_tcon(struct cifs_tcon *tcon) /* tc_count can never go negative */ WARN_ON(tcon->tc_count < 0); + list_del_init(&tcon->tcon_list); + spin_unlock(&cifs_tcp_ses_lock); + if (tcon->use_witness) { int rc; @@ -2171,9 +2222,6 @@ cifs_put_tcon(struct cifs_tcon *tcon) } } - list_del_init(&tcon->tcon_list); - spin_unlock(&cifs_tcp_ses_lock); - xid = get_xid(); if (ses->server->ops->tree_disconnect) ses->server->ops->tree_disconnect(xid, tcon); @@ -2290,10 +2338,6 @@ cifs_get_tcon(struct cifs_ses *ses, struct smb3_fs_context *ctx) } } - /* - * BB Do we need to wrap session_mutex around this TCon call and Unix - * SetFS as we do on SessSetup and reconnect? - */ xid = get_xid(); rc = ses->server->ops->tree_connect(xid, ses, ctx->UNC, tcon, ctx->local_nls); @@ -3029,12 +3073,15 @@ static int mount_get_conns(struct mount_ctx *mnt_ctx) * for just this mount. */ reset_cifs_unix_caps(xid, tcon, cifs_sb, ctx); + spin_lock(&cifs_tcp_ses_lock); if ((tcon->ses->server->tcpStatus == CifsNeedReconnect) && (le64_to_cpu(tcon->fsUnixInfo.Capability) & CIFS_UNIX_TRANSPORT_ENCRYPTION_MANDATORY_CAP)) { + spin_unlock(&cifs_tcp_ses_lock); rc = -EACCES; goto out; } + spin_unlock(&cifs_tcp_ses_lock); } else tcon->unix_ext = 0; /* server does not support them */ @@ -3069,7 +3116,8 @@ static int mount_get_conns(struct mount_ctx *mnt_ctx) * Inside cifs_fscache_get_super_cookie it checks * that we do not get super cookie twice. */ - cifs_fscache_get_super_cookie(tcon); + if (cifs_sb->mnt_cifs_flags & CIFS_MOUNT_FSCACHE) + cifs_fscache_get_super_cookie(tcon); out: mnt_ctx->server = server; @@ -3322,6 +3370,11 @@ static int is_path_remote(struct mount_ctx *mnt_ctx) rc = server->ops->is_path_accessible(xid, tcon, cifs_sb, full_path); +#ifdef CONFIG_CIFS_DFS_UPCALL + if (rc == -ENOENT && is_tcon_dfs(tcon)) + rc = cifs_dfs_query_info_nonascii_quirk(xid, tcon, cifs_sb, + full_path); +#endif if (rc != 0 && rc != -EREMOTE) { kfree(full_path); return rc; @@ -3709,8 +3762,6 @@ CIFSTCon(const unsigned int xid, struct cifs_ses *ses, if (rc == 0) { bool is_unicode; - tcon->tidStatus = CifsGood; - tcon->need_reconnect = false; tcon->tid = smb_buffer_response->Tid; bcc_ptr = pByteArea(smb_buffer_response); bytes_left = get_bcc(smb_buffer_response); @@ -3799,26 +3850,37 @@ cifs_umount(struct cifs_sb_info *cifs_sb) } int -cifs_negotiate_protocol(const unsigned int xid, struct cifs_ses *ses) +cifs_negotiate_protocol(const unsigned int xid, struct cifs_ses *ses, + struct TCP_Server_Info *server) { int rc = 0; - struct TCP_Server_Info *server = cifs_ses_server(ses); if (!server->ops->need_neg || !server->ops->negotiate) return -ENOSYS; /* only send once per connect */ - if (!server->ops->need_neg(server)) + spin_lock(&cifs_tcp_ses_lock); + if (!server->ops->need_neg(server) || + server->tcpStatus != CifsNeedNegotiate) { + spin_unlock(&cifs_tcp_ses_lock); return 0; + } + server->tcpStatus = CifsInNegotiate; + spin_unlock(&cifs_tcp_ses_lock); - rc = server->ops->negotiate(xid, ses); + rc = server->ops->negotiate(xid, ses, server); if (rc == 0) { - spin_lock(&GlobalMid_Lock); - if (server->tcpStatus == CifsNeedNegotiate) - server->tcpStatus = CifsGood; + spin_lock(&cifs_tcp_ses_lock); + if (server->tcpStatus == CifsInNegotiate) + server->tcpStatus = CifsNeedSessSetup; else rc = -EHOSTDOWN; - spin_unlock(&GlobalMid_Lock); + spin_unlock(&cifs_tcp_ses_lock); + } else { + spin_lock(&cifs_tcp_ses_lock); + if (server->tcpStatus == CifsInNegotiate) + server->tcpStatus = CifsNeedNegotiate; + spin_unlock(&cifs_tcp_ses_lock); } return rc; @@ -3826,12 +3888,26 @@ cifs_negotiate_protocol(const unsigned int xid, struct cifs_ses *ses) int cifs_setup_session(const unsigned int xid, struct cifs_ses *ses, + struct TCP_Server_Info *server, struct nls_table *nls_info) { int rc = -ENOSYS; - struct TCP_Server_Info *server = cifs_ses_server(ses); + bool is_binding = false; - if (!ses->binding) { + /* only send once per connect */ + spin_lock(&cifs_tcp_ses_lock); + if (server->tcpStatus != CifsNeedSessSetup) { + spin_unlock(&cifs_tcp_ses_lock); + return 0; + } + server->tcpStatus = CifsInSessSetup; + spin_unlock(&cifs_tcp_ses_lock); + + spin_lock(&ses->chan_lock); + is_binding = !CIFS_ALL_CHANS_NEED_RECONNECT(ses); + spin_unlock(&ses->chan_lock); + + if (!is_binding) { ses->capabilities = server->capabilities; if (!linuxExtEnabled) ses->capabilities &= (~server->vals->cap_unix); @@ -3849,10 +3925,26 @@ cifs_setup_session(const unsigned int xid, struct cifs_ses *ses, server->sec_mode, server->capabilities, server->timeAdj); if (server->ops->sess_setup) - rc = server->ops->sess_setup(xid, ses, nls_info); + rc = server->ops->sess_setup(xid, ses, server, nls_info); - if (rc) + if (rc) { cifs_server_dbg(VFS, "Send error in SessSetup = %d\n", rc); + spin_lock(&cifs_tcp_ses_lock); + if (server->tcpStatus == CifsInSessSetup) + server->tcpStatus = CifsNeedSessSetup; + spin_unlock(&cifs_tcp_ses_lock); + } else { + spin_lock(&cifs_tcp_ses_lock); + if (server->tcpStatus == CifsInSessSetup) + server->tcpStatus = CifsGood; + /* Even if one channel is active, session is in good state */ + ses->status = CifsGood; + spin_unlock(&cifs_tcp_ses_lock); + + spin_lock(&ses->chan_lock); + cifs_chan_clear_need_reconnect(ses, server); + spin_unlock(&ses->chan_lock); + } return rc; } @@ -4296,7 +4388,7 @@ static int tree_connect_dfs_target(const unsigned int xid, struct cifs_tcon *tco */ if (rc && server->current_fullpath != server->origin_fullpath) { server->current_fullpath = server->origin_fullpath; - cifs_ses_mark_for_reconnect(tcon->ses); + cifs_reconnect(tcon->ses->server, true); } dfs_cache_free_tgts(tl); @@ -4314,9 +4406,22 @@ int cifs_tree_connect(const unsigned int xid, struct cifs_tcon *tcon, const stru char *tree; struct dfs_info3_param ref = {0}; + /* only send once per connect */ + spin_lock(&cifs_tcp_ses_lock); + if (tcon->ses->status != CifsGood || + (tcon->tidStatus != CifsNew && + tcon->tidStatus != CifsNeedTcon)) { + spin_unlock(&cifs_tcp_ses_lock); + return 0; + } + tcon->tidStatus = CifsInTcon; + spin_unlock(&cifs_tcp_ses_lock); + tree = kzalloc(MAX_TREE_SIZE, GFP_KERNEL); - if (!tree) - return -ENOMEM; + if (!tree) { + rc = -ENOMEM; + goto out; + } if (tcon->ipc) { scnprintf(tree, MAX_TREE_SIZE, "\\\\%s\\IPC$", server->hostname); @@ -4348,13 +4453,52 @@ out: kfree(tree); cifs_put_tcp_super(sb); + if (rc) { + spin_lock(&cifs_tcp_ses_lock); + if (tcon->tidStatus == CifsInTcon) + tcon->tidStatus = CifsNeedTcon; + spin_unlock(&cifs_tcp_ses_lock); + } else { + spin_lock(&cifs_tcp_ses_lock); + if (tcon->tidStatus == CifsInTcon) + tcon->tidStatus = CifsGood; + spin_unlock(&cifs_tcp_ses_lock); + tcon->need_reconnect = false; + } + return rc; } #else int cifs_tree_connect(const unsigned int xid, struct cifs_tcon *tcon, const struct nls_table *nlsc) { + int rc; const struct smb_version_operations *ops = tcon->ses->server->ops; - return ops->tree_connect(xid, tcon->ses, tcon->treeName, tcon, nlsc); + /* only send once per connect */ + spin_lock(&cifs_tcp_ses_lock); + if (tcon->ses->status != CifsGood || + (tcon->tidStatus != CifsNew && + tcon->tidStatus != CifsNeedTcon)) { + spin_unlock(&cifs_tcp_ses_lock); + return 0; + } + tcon->tidStatus = CifsInTcon; + spin_unlock(&cifs_tcp_ses_lock); + + rc = ops->tree_connect(xid, tcon->ses, tcon->treeName, tcon, nlsc); + if (rc) { + spin_lock(&cifs_tcp_ses_lock); + if (tcon->tidStatus == CifsInTcon) + tcon->tidStatus = CifsNeedTcon; + spin_unlock(&cifs_tcp_ses_lock); + } else { + spin_lock(&cifs_tcp_ses_lock); + if (tcon->tidStatus == CifsInTcon) + tcon->tidStatus = CifsGood; + spin_unlock(&cifs_tcp_ses_lock); + tcon->need_reconnect = false; + } + + return rc; } #endif diff --git a/fs/cifs/dfs_cache.c b/fs/cifs/dfs_cache.c index e9b0fa2a9614..dd9643751671 100644 --- a/fs/cifs/dfs_cache.c +++ b/fs/cifs/dfs_cache.c @@ -1355,7 +1355,7 @@ static void mark_for_reconnect_if_needed(struct cifs_tcon *tcon, struct dfs_cach } cifs_dbg(FYI, "%s: no cached or matched targets. mark dfs share for reconnect.\n", __func__); - cifs_ses_mark_for_reconnect(tcon->ses); + cifs_reconnect(tcon->ses->server, true); } /* Refresh dfs referral of tcon and mark it for reconnect if needed */ diff --git a/fs/cifs/dir.c b/fs/cifs/dir.c index 6e8e7cc26ae2..ce9b22aecfba 100644 --- a/fs/cifs/dir.c +++ b/fs/cifs/dir.c @@ -22,6 +22,7 @@ #include "cifs_unicode.h" #include "fs_context.h" #include "cifs_ioctl.h" +#include "fscache.h" static void renew_parental_timestamps(struct dentry *direntry) @@ -507,8 +508,12 @@ cifs_atomic_open(struct inode *inode, struct dentry *direntry, server->ops->close(xid, tcon, &fid); cifs_del_pending_open(&open); rc = -ENOMEM; + goto out; } + fscache_use_cookie(cifs_inode_cookie(file_inode(file)), + file->f_mode & FMODE_WRITE); + out: cifs_put_tlink(tlink); out_free_xid: diff --git a/fs/cifs/file.c b/fs/cifs/file.c index 9fee3af83a73..59334be9ed3b 100644 --- a/fs/cifs/file.c +++ b/fs/cifs/file.c @@ -376,8 +376,6 @@ static void cifsFileInfo_put_final(struct cifsFileInfo *cifs_file) struct cifsLockInfo *li, *tmp; struct super_block *sb = inode->i_sb; - cifs_fscache_release_inode_cookie(inode); - /* * Delete any outstanding lock records. We'll lose them when the file * is closed anyway. @@ -570,7 +568,7 @@ int cifs_open(struct inode *inode, struct file *file) spin_lock(&CIFS_I(inode)->deferred_lock); cifs_del_deferred_close(cfile); spin_unlock(&CIFS_I(inode)->deferred_lock); - goto out; + goto use_cache; } else { _cifsFileInfo_put(cfile, true, false); } @@ -632,8 +630,6 @@ int cifs_open(struct inode *inode, struct file *file) goto out; } - cifs_fscache_set_inode_cookie(inode, file); - if ((oplock & CIFS_CREATE_ACTION) && !posix_open_ok && tcon->unix_ext) { /* * Time to set mode which we can not set earlier due to @@ -652,6 +648,15 @@ int cifs_open(struct inode *inode, struct file *file) cfile->pid); } +use_cache: + fscache_use_cookie(cifs_inode_cookie(file_inode(file)), + file->f_mode & FMODE_WRITE); + if (file->f_flags & O_DIRECT && + (!((file->f_flags & O_ACCMODE) != O_RDONLY) || + file->f_flags & O_APPEND)) + cifs_invalidate_cache(file_inode(file), + FSCACHE_INVAL_DIO_WRITE); + out: free_dentry_path(page); free_xid(xid); @@ -876,6 +881,8 @@ int cifs_close(struct inode *inode, struct file *file) struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); struct cifs_deferred_close *dclose; + cifs_fscache_unuse_inode_cookie(inode, file->f_mode & FMODE_WRITE); + if (file->private_data != NULL) { cfile = file->private_data; file->private_data = NULL; @@ -886,7 +893,6 @@ int cifs_close(struct inode *inode, struct file *file) dclose) { if (test_and_clear_bit(CIFS_INO_MODIFIED_ATTR, &cinode->flags)) { inode->i_ctime = inode->i_mtime = current_time(inode); - cifs_fscache_update_inode_cookie(inode); } spin_lock(&cinode->deferred_lock); cifs_add_deferred_close(cfile, dclose); @@ -4198,10 +4204,12 @@ static vm_fault_t cifs_page_mkwrite(struct vm_fault *vmf) { struct page *page = vmf->page; - struct file *file = vmf->vma->vm_file; - struct inode *inode = file_inode(file); - cifs_fscache_wait_on_page_write(inode, page); +#ifdef CONFIG_CIFS_FSCACHE + if (PageFsCache(page) && + wait_on_page_fscache_killable(page) < 0) + return VM_FAULT_RETRY; +#endif lock_page(page); return VM_FAULT_LOCKED; @@ -4275,8 +4283,6 @@ cifs_readv_complete(struct work_struct *work) if (rdata->result == 0 || (rdata->result == -EAGAIN && got_bytes)) cifs_readpage_to_fscache(rdata->mapping->host, page); - else - cifs_fscache_uncache_page(rdata->mapping->host, page); got_bytes -= min_t(unsigned int, PAGE_SIZE, got_bytes); @@ -4593,11 +4599,6 @@ static int cifs_readpages(struct file *file, struct address_space *mapping, kref_put(&rdata->refcount, cifs_readdata_release); } - /* Any pages that have been shown to fscache but didn't get added to - * the pagecache must be uncached before they get returned to the - * allocator. - */ - cifs_fscache_readpages_cancel(mapping->host, page_list); free_xid(xid); return rc; } @@ -4801,17 +4802,19 @@ static int cifs_release_page(struct page *page, gfp_t gfp) { if (PagePrivate(page)) return 0; - - return cifs_fscache_release_page(page, gfp); + if (PageFsCache(page)) { + if (current_is_kswapd() || !(gfp & __GFP_FS)) + return false; + wait_on_page_fscache(page); + } + fscache_note_page_release(cifs_inode_cookie(page->mapping->host)); + return true; } static void cifs_invalidate_page(struct page *page, unsigned int offset, unsigned int length) { - struct cifsInodeInfo *cifsi = CIFS_I(page->mapping->host); - - if (offset == 0 && length == PAGE_SIZE) - cifs_fscache_invalidate_page(page, &cifsi->vfs_inode); + wait_on_page_fscache(page); } static int cifs_launder_page(struct page *page) @@ -4831,7 +4834,7 @@ static int cifs_launder_page(struct page *page) if (clear_page_dirty_for_io(page)) rc = cifs_writepage_locked(page, &wbc); - cifs_fscache_invalidate_page(page, page->mapping->host); + wait_on_page_fscache(page); return rc; } @@ -4988,6 +4991,19 @@ static void cifs_swap_deactivate(struct file *file) /* do we need to unpin (or unlock) the file */ } +/* + * Mark a page as having been made dirty and thus needing writeback. We also + * need to pin the cache object to write back to. + */ +#ifdef CONFIG_CIFS_FSCACHE +static int cifs_set_page_dirty(struct page *page) +{ + return fscache_set_page_dirty(page, cifs_inode_cookie(page->mapping->host)); +} +#else +#define cifs_set_page_dirty __set_page_dirty_nobuffers +#endif + const struct address_space_operations cifs_addr_ops = { .readpage = cifs_readpage, .readpages = cifs_readpages, @@ -4995,7 +5011,7 @@ const struct address_space_operations cifs_addr_ops = { .writepages = cifs_writepages, .write_begin = cifs_write_begin, .write_end = cifs_write_end, - .set_page_dirty = __set_page_dirty_nobuffers, + .set_page_dirty = cifs_set_page_dirty, .releasepage = cifs_release_page, .direct_IO = cifs_direct_io, .invalidatepage = cifs_invalidate_page, @@ -5020,7 +5036,7 @@ const struct address_space_operations cifs_addr_ops_smallbuf = { .writepages = cifs_writepages, .write_begin = cifs_write_begin, .write_end = cifs_write_end, - .set_page_dirty = __set_page_dirty_nobuffers, + .set_page_dirty = cifs_set_page_dirty, .releasepage = cifs_release_page, .invalidatepage = cifs_invalidate_page, .launder_page = cifs_launder_page, diff --git a/fs/cifs/fs_context.c b/fs/cifs/fs_context.c index e3ed25dc6f3f..7ec35f3f0a5f 100644 --- a/fs/cifs/fs_context.c +++ b/fs/cifs/fs_context.c @@ -37,6 +37,8 @@ #include "rfc1002pdu.h" #include "fs_context.h" +static DEFINE_MUTEX(cifs_mount_mutex); + static const match_table_t cifs_smb_version_tokens = { { Smb_1, SMB1_VERSION_STRING }, { Smb_20, SMB20_VERSION_STRING}, @@ -707,10 +709,14 @@ static int smb3_get_tree_common(struct fs_context *fc) static int smb3_get_tree(struct fs_context *fc) { int err = smb3_fs_context_validate(fc); + int ret; if (err) return err; - return smb3_get_tree_common(fc); + mutex_lock(&cifs_mount_mutex); + ret = smb3_get_tree_common(fc); + mutex_unlock(&cifs_mount_mutex); + return ret; } static void smb3_fs_context_free(struct fs_context *fc) diff --git a/fs/cifs/fscache.c b/fs/cifs/fscache.c index 003c5f1f4dfb..efaac4d5ff55 100644 --- a/fs/cifs/fscache.c +++ b/fs/cifs/fscache.c @@ -12,250 +12,136 @@ #include "cifs_fs_sb.h" #include "cifsproto.h" -/* - * Key layout of CIFS server cache index object - */ -struct cifs_server_key { - __u64 conn_id; -} __packed; - -/* - * Get a cookie for a server object keyed by {IPaddress,port,family} tuple - */ -void cifs_fscache_get_client_cookie(struct TCP_Server_Info *server) -{ - struct cifs_server_key key; - - /* - * Check if cookie was already initialized so don't reinitialize it. - * In the future, as we integrate with newer fscache features, - * we may want to instead add a check if cookie has changed - */ - if (server->fscache) - return; - - memset(&key, 0, sizeof(key)); - key.conn_id = server->conn_id; - - server->fscache = - fscache_acquire_cookie(cifs_fscache_netfs.primary_index, - &cifs_fscache_server_index_def, - &key, sizeof(key), - NULL, 0, - server, 0, true); - cifs_dbg(FYI, "%s: (0x%p/0x%p)\n", - __func__, server, server->fscache); -} - -void cifs_fscache_release_client_cookie(struct TCP_Server_Info *server) +static void cifs_fscache_fill_volume_coherency( + struct cifs_tcon *tcon, + struct cifs_fscache_volume_coherency_data *cd) { - cifs_dbg(FYI, "%s: (0x%p/0x%p)\n", - __func__, server, server->fscache); - fscache_relinquish_cookie(server->fscache, NULL, false); - server->fscache = NULL; + memset(cd, 0, sizeof(*cd)); + cd->resource_id = cpu_to_le64(tcon->resource_id); + cd->vol_create_time = tcon->vol_create_time; + cd->vol_serial_number = cpu_to_le32(tcon->vol_serial_number); } -void cifs_fscache_get_super_cookie(struct cifs_tcon *tcon) +int cifs_fscache_get_super_cookie(struct cifs_tcon *tcon) { + struct cifs_fscache_volume_coherency_data cd; struct TCP_Server_Info *server = tcon->ses->server; + struct fscache_volume *vcookie; + const struct sockaddr *sa = (struct sockaddr *)&server->dstaddr; + size_t slen, i; char *sharename; - struct cifs_fscache_super_auxdata auxdata; + char *key; + int ret = -ENOMEM; - /* - * Check if cookie was already initialized so don't reinitialize it. - * In the future, as we integrate with newer fscache features, - * we may want to instead add a check if cookie has changed - */ - if (tcon->fscache) - return; + tcon->fscache = NULL; + switch (sa->sa_family) { + case AF_INET: + case AF_INET6: + break; + default: + cifs_dbg(VFS, "Unknown network family '%d'\n", sa->sa_family); + return -EINVAL; + } + + memset(&key, 0, sizeof(key)); sharename = extract_sharename(tcon->treeName); if (IS_ERR(sharename)) { cifs_dbg(FYI, "%s: couldn't extract sharename\n", __func__); - tcon->fscache = NULL; - return; + return -EINVAL; } - memset(&auxdata, 0, sizeof(auxdata)); - auxdata.resource_id = tcon->resource_id; - auxdata.vol_create_time = tcon->vol_create_time; - auxdata.vol_serial_number = tcon->vol_serial_number; + slen = strlen(sharename); + for (i = 0; i < slen; i++) + if (sharename[i] == '/') + sharename[i] = ';'; + + key = kasprintf(GFP_KERNEL, "cifs,%pISpc,%s", sa, sharename); + if (!key) + goto out; + + cifs_fscache_fill_volume_coherency(tcon, &cd); + vcookie = fscache_acquire_volume(key, + NULL, /* preferred_cache */ + &cd, sizeof(cd)); + cifs_dbg(FYI, "%s: (%s/0x%p)\n", __func__, key, vcookie); + if (IS_ERR(vcookie)) { + if (vcookie != ERR_PTR(-EBUSY)) { + ret = PTR_ERR(vcookie); + goto out_2; + } + pr_err("Cache volume key already in use (%s)\n", key); + vcookie = NULL; + } - tcon->fscache = - fscache_acquire_cookie(server->fscache, - &cifs_fscache_super_index_def, - sharename, strlen(sharename), - &auxdata, sizeof(auxdata), - tcon, 0, true); + tcon->fscache = vcookie; + ret = 0; +out_2: + kfree(key); +out: kfree(sharename); - cifs_dbg(FYI, "%s: (0x%p/0x%p)\n", - __func__, server->fscache, tcon->fscache); + return ret; } void cifs_fscache_release_super_cookie(struct cifs_tcon *tcon) { - struct cifs_fscache_super_auxdata auxdata; - - memset(&auxdata, 0, sizeof(auxdata)); - auxdata.resource_id = tcon->resource_id; - auxdata.vol_create_time = tcon->vol_create_time; - auxdata.vol_serial_number = tcon->vol_serial_number; + struct cifs_fscache_volume_coherency_data cd; cifs_dbg(FYI, "%s: (0x%p)\n", __func__, tcon->fscache); - fscache_relinquish_cookie(tcon->fscache, &auxdata, false); - tcon->fscache = NULL; -} - -static void cifs_fscache_acquire_inode_cookie(struct cifsInodeInfo *cifsi, - struct cifs_tcon *tcon) -{ - struct cifs_fscache_inode_auxdata auxdata; - memset(&auxdata, 0, sizeof(auxdata)); - auxdata.eof = cifsi->server_eof; - auxdata.last_write_time_sec = cifsi->vfs_inode.i_mtime.tv_sec; - auxdata.last_change_time_sec = cifsi->vfs_inode.i_ctime.tv_sec; - auxdata.last_write_time_nsec = cifsi->vfs_inode.i_mtime.tv_nsec; - auxdata.last_change_time_nsec = cifsi->vfs_inode.i_ctime.tv_nsec; - - cifsi->fscache = - fscache_acquire_cookie(tcon->fscache, - &cifs_fscache_inode_object_def, - &cifsi->uniqueid, sizeof(cifsi->uniqueid), - &auxdata, sizeof(auxdata), - cifsi, cifsi->vfs_inode.i_size, true); + cifs_fscache_fill_volume_coherency(tcon, &cd); + fscache_relinquish_volume(tcon->fscache, &cd, false); + tcon->fscache = NULL; } -static void cifs_fscache_enable_inode_cookie(struct inode *inode) +void cifs_fscache_get_inode_cookie(struct inode *inode) { + struct cifs_fscache_inode_coherency_data cd; struct cifsInodeInfo *cifsi = CIFS_I(inode); struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); struct cifs_tcon *tcon = cifs_sb_master_tcon(cifs_sb); - if (cifsi->fscache) - return; - - if (!(cifs_sb->mnt_cifs_flags & CIFS_MOUNT_FSCACHE)) - return; - - cifs_fscache_acquire_inode_cookie(cifsi, tcon); + cifs_fscache_fill_coherency(&cifsi->vfs_inode, &cd); - cifs_dbg(FYI, "%s: got FH cookie (0x%p/0x%p)\n", - __func__, tcon->fscache, cifsi->fscache); + cifsi->fscache = + fscache_acquire_cookie(tcon->fscache, 0, + &cifsi->uniqueid, sizeof(cifsi->uniqueid), + &cd, sizeof(cd), + i_size_read(&cifsi->vfs_inode)); } -void cifs_fscache_release_inode_cookie(struct inode *inode) +void cifs_fscache_unuse_inode_cookie(struct inode *inode, bool update) { - struct cifs_fscache_inode_auxdata auxdata; - struct cifsInodeInfo *cifsi = CIFS_I(inode); - - if (cifsi->fscache) { - memset(&auxdata, 0, sizeof(auxdata)); - auxdata.eof = cifsi->server_eof; - auxdata.last_write_time_sec = cifsi->vfs_inode.i_mtime.tv_sec; - auxdata.last_change_time_sec = cifsi->vfs_inode.i_ctime.tv_sec; - auxdata.last_write_time_nsec = cifsi->vfs_inode.i_mtime.tv_nsec; - auxdata.last_change_time_nsec = cifsi->vfs_inode.i_ctime.tv_nsec; + if (update) { + struct cifs_fscache_inode_coherency_data cd; + loff_t i_size = i_size_read(inode); - cifs_dbg(FYI, "%s: (0x%p)\n", __func__, cifsi->fscache); - /* fscache_relinquish_cookie does not seem to update auxdata */ - fscache_update_cookie(cifsi->fscache, &auxdata); - fscache_relinquish_cookie(cifsi->fscache, &auxdata, false); - cifsi->fscache = NULL; + cifs_fscache_fill_coherency(inode, &cd); + fscache_unuse_cookie(cifs_inode_cookie(inode), &cd, &i_size); + } else { + fscache_unuse_cookie(cifs_inode_cookie(inode), NULL, NULL); } } -void cifs_fscache_update_inode_cookie(struct inode *inode) +void cifs_fscache_release_inode_cookie(struct inode *inode) { - struct cifs_fscache_inode_auxdata auxdata; struct cifsInodeInfo *cifsi = CIFS_I(inode); if (cifsi->fscache) { - memset(&auxdata, 0, sizeof(auxdata)); - auxdata.eof = cifsi->server_eof; - auxdata.last_write_time_sec = cifsi->vfs_inode.i_mtime.tv_sec; - auxdata.last_change_time_sec = cifsi->vfs_inode.i_ctime.tv_sec; - auxdata.last_write_time_nsec = cifsi->vfs_inode.i_mtime.tv_nsec; - auxdata.last_change_time_nsec = cifsi->vfs_inode.i_ctime.tv_nsec; - cifs_dbg(FYI, "%s: (0x%p)\n", __func__, cifsi->fscache); - fscache_update_cookie(cifsi->fscache, &auxdata); - } -} - -void cifs_fscache_set_inode_cookie(struct inode *inode, struct file *filp) -{ - cifs_fscache_enable_inode_cookie(inode); -} - -void cifs_fscache_reset_inode_cookie(struct inode *inode) -{ - struct cifsInodeInfo *cifsi = CIFS_I(inode); - struct cifs_sb_info *cifs_sb = CIFS_SB(inode->i_sb); - struct cifs_tcon *tcon = cifs_sb_master_tcon(cifs_sb); - struct fscache_cookie *old = cifsi->fscache; - - if (cifsi->fscache) { - /* retire the current fscache cache and get a new one */ - fscache_relinquish_cookie(cifsi->fscache, NULL, true); - - cifs_fscache_acquire_inode_cookie(cifsi, tcon); - cifs_dbg(FYI, "%s: new cookie 0x%p oldcookie 0x%p\n", - __func__, cifsi->fscache, old); + fscache_relinquish_cookie(cifsi->fscache, false); + cifsi->fscache = NULL; } } -int cifs_fscache_release_page(struct page *page, gfp_t gfp) -{ - if (PageFsCache(page)) { - struct inode *inode = page->mapping->host; - struct cifsInodeInfo *cifsi = CIFS_I(inode); - - cifs_dbg(FYI, "%s: (0x%p/0x%p)\n", - __func__, page, cifsi->fscache); - if (!fscache_maybe_release_page(cifsi->fscache, page, gfp)) - return 0; - } - - return 1; -} - -static void cifs_readpage_from_fscache_complete(struct page *page, void *ctx, - int error) -{ - cifs_dbg(FYI, "%s: (0x%p/%d)\n", __func__, page, error); - if (!error) - SetPageUptodate(page); - unlock_page(page); -} - /* * Retrieve a page from FS-Cache */ int __cifs_readpage_from_fscache(struct inode *inode, struct page *page) { - int ret; - cifs_dbg(FYI, "%s: (fsc:%p, p:%p, i:0x%p\n", __func__, CIFS_I(inode)->fscache, page, inode); - ret = fscache_read_or_alloc_page(CIFS_I(inode)->fscache, page, - cifs_readpage_from_fscache_complete, - NULL, - GFP_KERNEL); - switch (ret) { - - case 0: /* page found in fscache, read submitted */ - cifs_dbg(FYI, "%s: submitted\n", __func__); - return ret; - case -ENOBUFS: /* page won't be cached */ - case -ENODATA: /* page not in cache */ - cifs_dbg(FYI, "%s: %d\n", __func__, ret); - return 1; - - default: - cifs_dbg(VFS, "unknown error ret = %d\n", ret); - } - return ret; + return -ENOBUFS; // Needs conversion to using netfslib } /* @@ -266,78 +152,19 @@ int __cifs_readpages_from_fscache(struct inode *inode, struct list_head *pages, unsigned *nr_pages) { - int ret; - cifs_dbg(FYI, "%s: (0x%p/%u/0x%p)\n", __func__, CIFS_I(inode)->fscache, *nr_pages, inode); - ret = fscache_read_or_alloc_pages(CIFS_I(inode)->fscache, mapping, - pages, nr_pages, - cifs_readpage_from_fscache_complete, - NULL, - mapping_gfp_mask(mapping)); - switch (ret) { - case 0: /* read submitted to the cache for all pages */ - cifs_dbg(FYI, "%s: submitted\n", __func__); - return ret; - - case -ENOBUFS: /* some pages are not cached and can't be */ - case -ENODATA: /* some pages are not cached */ - cifs_dbg(FYI, "%s: no page\n", __func__); - return 1; - - default: - cifs_dbg(FYI, "unknown error ret = %d\n", ret); - } - - return ret; + return -ENOBUFS; // Needs conversion to using netfslib } void __cifs_readpage_to_fscache(struct inode *inode, struct page *page) { struct cifsInodeInfo *cifsi = CIFS_I(inode); - int ret; WARN_ON(!cifsi->fscache); cifs_dbg(FYI, "%s: (fsc: %p, p: %p, i: %p)\n", __func__, cifsi->fscache, page, inode); - ret = fscache_write_page(cifsi->fscache, page, - cifsi->vfs_inode.i_size, GFP_KERNEL); - if (ret != 0) - fscache_uncache_page(cifsi->fscache, page); -} - -void __cifs_fscache_readpages_cancel(struct inode *inode, struct list_head *pages) -{ - cifs_dbg(FYI, "%s: (fsc: %p, i: %p)\n", - __func__, CIFS_I(inode)->fscache, inode); - fscache_readpages_cancel(CIFS_I(inode)->fscache, pages); -} - -void __cifs_fscache_invalidate_page(struct page *page, struct inode *inode) -{ - struct cifsInodeInfo *cifsi = CIFS_I(inode); - struct fscache_cookie *cookie = cifsi->fscache; - - cifs_dbg(FYI, "%s: (0x%p/0x%p)\n", __func__, page, cookie); - fscache_wait_on_page_write(cookie, page); - fscache_uncache_page(cookie, page); -} - -void __cifs_fscache_wait_on_page_write(struct inode *inode, struct page *page) -{ - struct cifsInodeInfo *cifsi = CIFS_I(inode); - struct fscache_cookie *cookie = cifsi->fscache; - - cifs_dbg(FYI, "%s: (0x%p/0x%p)\n", __func__, page, cookie); - fscache_wait_on_page_write(cookie, page); -} - -void __cifs_fscache_uncache_page(struct inode *inode, struct page *page) -{ - struct cifsInodeInfo *cifsi = CIFS_I(inode); - struct fscache_cookie *cookie = cifsi->fscache; - cifs_dbg(FYI, "%s: (0x%p/0x%p)\n", __func__, page, cookie); - fscache_uncache_page(cookie, page); + // Needs conversion to using netfslib } diff --git a/fs/cifs/fscache.h b/fs/cifs/fscache.h index 9baa1d0f22bd..c6ca49ac33d4 100644 --- a/fs/cifs/fscache.h +++ b/fs/cifs/fscache.h @@ -13,84 +13,71 @@ #include "cifsglob.h" -#ifdef CONFIG_CIFS_FSCACHE - /* - * Auxiliary data attached to CIFS superblock within the cache + * Coherency data attached to CIFS volume within the cache */ -struct cifs_fscache_super_auxdata { - u64 resource_id; /* unique server resource id */ +struct cifs_fscache_volume_coherency_data { + __le64 resource_id; /* unique server resource id */ __le64 vol_create_time; - u32 vol_serial_number; + __le32 vol_serial_number; } __packed; /* - * Auxiliary data attached to CIFS inode within the cache + * Coherency data attached to CIFS inode within the cache. */ -struct cifs_fscache_inode_auxdata { - u64 last_write_time_sec; - u64 last_change_time_sec; - u32 last_write_time_nsec; - u32 last_change_time_nsec; - u64 eof; +struct cifs_fscache_inode_coherency_data { + __le64 last_write_time_sec; + __le64 last_change_time_sec; + __le32 last_write_time_nsec; + __le32 last_change_time_nsec; }; -/* - * cache.c - */ -extern struct fscache_netfs cifs_fscache_netfs; -extern const struct fscache_cookie_def cifs_fscache_server_index_def; -extern const struct fscache_cookie_def cifs_fscache_super_index_def; -extern const struct fscache_cookie_def cifs_fscache_inode_object_def; - -extern int cifs_fscache_register(void); -extern void cifs_fscache_unregister(void); +#ifdef CONFIG_CIFS_FSCACHE /* * fscache.c */ -extern void cifs_fscache_get_client_cookie(struct TCP_Server_Info *); -extern void cifs_fscache_release_client_cookie(struct TCP_Server_Info *); -extern void cifs_fscache_get_super_cookie(struct cifs_tcon *); +extern int cifs_fscache_get_super_cookie(struct cifs_tcon *); extern void cifs_fscache_release_super_cookie(struct cifs_tcon *); +extern void cifs_fscache_get_inode_cookie(struct inode *inode); extern void cifs_fscache_release_inode_cookie(struct inode *); -extern void cifs_fscache_update_inode_cookie(struct inode *inode); -extern void cifs_fscache_set_inode_cookie(struct inode *, struct file *); -extern void cifs_fscache_reset_inode_cookie(struct inode *); +extern void cifs_fscache_unuse_inode_cookie(struct inode *inode, bool update); + +static inline +void cifs_fscache_fill_coherency(struct inode *inode, + struct cifs_fscache_inode_coherency_data *cd) +{ + struct cifsInodeInfo *cifsi = CIFS_I(inode); + + memset(cd, 0, sizeof(*cd)); + cd->last_write_time_sec = cpu_to_le64(cifsi->vfs_inode.i_mtime.tv_sec); + cd->last_write_time_nsec = cpu_to_le32(cifsi->vfs_inode.i_mtime.tv_nsec); + cd->last_change_time_sec = cpu_to_le64(cifsi->vfs_inode.i_ctime.tv_sec); + cd->last_change_time_nsec = cpu_to_le32(cifsi->vfs_inode.i_ctime.tv_nsec); +} + -extern void __cifs_fscache_invalidate_page(struct page *, struct inode *); -extern void __cifs_fscache_wait_on_page_write(struct inode *inode, struct page *page); -extern void __cifs_fscache_uncache_page(struct inode *inode, struct page *page); extern int cifs_fscache_release_page(struct page *page, gfp_t gfp); extern int __cifs_readpage_from_fscache(struct inode *, struct page *); extern int __cifs_readpages_from_fscache(struct inode *, struct address_space *, struct list_head *, unsigned *); -extern void __cifs_fscache_readpages_cancel(struct inode *, struct list_head *); - extern void __cifs_readpage_to_fscache(struct inode *, struct page *); -static inline void cifs_fscache_invalidate_page(struct page *page, - struct inode *inode) +static inline struct fscache_cookie *cifs_inode_cookie(struct inode *inode) { - if (PageFsCache(page)) - __cifs_fscache_invalidate_page(page, inode); + return CIFS_I(inode)->fscache; } -static inline void cifs_fscache_wait_on_page_write(struct inode *inode, - struct page *page) +static inline void cifs_invalidate_cache(struct inode *inode, unsigned int flags) { - if (PageFsCache(page)) - __cifs_fscache_wait_on_page_write(inode, page); -} + struct cifs_fscache_inode_coherency_data cd; -static inline void cifs_fscache_uncache_page(struct inode *inode, - struct page *page) -{ - if (PageFsCache(page)) - __cifs_fscache_uncache_page(inode, page); + cifs_fscache_fill_coherency(inode, &cd); + fscache_invalidate(cifs_inode_cookie(inode), &cd, + i_size_read(inode), flags); } static inline int cifs_readpage_from_fscache(struct inode *inode, @@ -120,41 +107,21 @@ static inline void cifs_readpage_to_fscache(struct inode *inode, __cifs_readpage_to_fscache(inode, page); } -static inline void cifs_fscache_readpages_cancel(struct inode *inode, - struct list_head *pages) +#else /* CONFIG_CIFS_FSCACHE */ +static inline +void cifs_fscache_fill_coherency(struct inode *inode, + struct cifs_fscache_inode_coherency_data *cd) { - if (CIFS_I(inode)->fscache) - return __cifs_fscache_readpages_cancel(inode, pages); } -#else /* CONFIG_CIFS_FSCACHE */ -static inline int cifs_fscache_register(void) { return 0; } -static inline void cifs_fscache_unregister(void) {} - -static inline void -cifs_fscache_get_client_cookie(struct TCP_Server_Info *server) {} -static inline void -cifs_fscache_release_client_cookie(struct TCP_Server_Info *server) {} -static inline void cifs_fscache_get_super_cookie(struct cifs_tcon *tcon) {} -static inline void -cifs_fscache_release_super_cookie(struct cifs_tcon *tcon) {} +static inline int cifs_fscache_get_super_cookie(struct cifs_tcon *tcon) { return 0; } +static inline void cifs_fscache_release_super_cookie(struct cifs_tcon *tcon) {} +static inline void cifs_fscache_get_inode_cookie(struct inode *inode) {} static inline void cifs_fscache_release_inode_cookie(struct inode *inode) {} -static inline void cifs_fscache_update_inode_cookie(struct inode *inode) {} -static inline void cifs_fscache_set_inode_cookie(struct inode *inode, - struct file *filp) {} -static inline void cifs_fscache_reset_inode_cookie(struct inode *inode) {} -static inline int cifs_fscache_release_page(struct page *page, gfp_t gfp) -{ - return 1; /* May release page */ -} - -static inline void cifs_fscache_invalidate_page(struct page *page, - struct inode *inode) {} -static inline void cifs_fscache_wait_on_page_write(struct inode *inode, - struct page *page) {} -static inline void cifs_fscache_uncache_page(struct inode *inode, - struct page *page) {} +static inline void cifs_fscache_unuse_inode_cookie(struct inode *inode, bool update) {} +static inline struct fscache_cookie *cifs_inode_cookie(struct inode *inode) { return NULL; } +static inline void cifs_invalidate_cache(struct inode *inode, unsigned int flags) {} static inline int cifs_readpage_from_fscache(struct inode *inode, struct page *page) @@ -173,11 +140,6 @@ static inline int cifs_readpages_from_fscache(struct inode *inode, static inline void cifs_readpage_to_fscache(struct inode *inode, struct page *page) {} -static inline void cifs_fscache_readpages_cancel(struct inode *inode, - struct list_head *pages) -{ -} - #endif /* CONFIG_CIFS_FSCACHE */ #endif /* _CIFS_FSCACHE_H */ diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c index 279622e4eb1c..7d8b3ceb2af3 100644 --- a/fs/cifs/inode.c +++ b/fs/cifs/inode.c @@ -952,6 +952,12 @@ cifs_get_inode_info(struct inode **inode, rc = server->ops->query_path_info(xid, tcon, cifs_sb, full_path, tmp_data, &adjust_tz, &is_reparse_point); +#ifdef CONFIG_CIFS_DFS_UPCALL + if (rc == -ENOENT && is_tcon_dfs(tcon)) + rc = cifs_dfs_query_info_nonascii_quirk(xid, tcon, + cifs_sb, + full_path); +#endif data = tmp_data; } @@ -1298,10 +1304,7 @@ retry_iget5_locked: inode->i_flags |= S_NOATIME | S_NOCMTIME; if (inode->i_state & I_NEW) { inode->i_ino = hash; -#ifdef CONFIG_CIFS_FSCACHE - /* initialize per-inode cache cookie pointer */ - CIFS_I(inode)->fscache = NULL; -#endif + cifs_fscache_get_inode_cookie(inode); unlock_new_inode(inode); } } @@ -1370,6 +1373,7 @@ iget_no_retry: iget_failed(inode); inode = ERR_PTR(rc); } + out: kfree(path); free_xid(xid); @@ -2257,6 +2261,8 @@ cifs_dentry_needs_reval(struct dentry *dentry) int cifs_invalidate_mapping(struct inode *inode) { + struct cifs_fscache_inode_coherency_data cd; + struct cifsInodeInfo *cifsi = CIFS_I(inode); int rc = 0; if (inode->i_mapping && inode->i_mapping->nrpages != 0) { @@ -2266,7 +2272,8 @@ cifs_invalidate_mapping(struct inode *inode) __func__, inode); } - cifs_fscache_reset_inode_cookie(inode); + cifs_fscache_fill_coherency(&cifsi->vfs_inode, &cd); + fscache_invalidate(cifs_inode_cookie(inode), &cd, i_size_read(inode), 0); return rc; } @@ -2771,8 +2778,10 @@ cifs_setattr_unix(struct dentry *direntry, struct iattr *attrs) goto out; if ((attrs->ia_valid & ATTR_SIZE) && - attrs->ia_size != i_size_read(inode)) + attrs->ia_size != i_size_read(inode)) { truncate_setsize(inode, attrs->ia_size); + fscache_resize_cookie(cifs_inode_cookie(inode), attrs->ia_size); + } setattr_copy(&init_user_ns, inode, attrs); mark_inode_dirty(inode); @@ -2967,8 +2976,10 @@ cifs_setattr_nounix(struct dentry *direntry, struct iattr *attrs) goto cifs_setattr_exit; if ((attrs->ia_valid & ATTR_SIZE) && - attrs->ia_size != i_size_read(inode)) + attrs->ia_size != i_size_read(inode)) { truncate_setsize(inode, attrs->ia_size); + fscache_resize_cookie(cifs_inode_cookie(inode), attrs->ia_size); + } setattr_copy(&init_user_ns, inode, attrs); mark_inode_dirty(inode); diff --git a/fs/cifs/misc.c b/fs/cifs/misc.c index 5148d48d6a35..56598f7dbe00 100644 --- a/fs/cifs/misc.c +++ b/fs/cifs/misc.c @@ -1302,4 +1302,53 @@ int cifs_update_super_prepath(struct cifs_sb_info *cifs_sb, char *prefix) cifs_sb->mnt_cifs_flags |= CIFS_MOUNT_USE_PREFIX_PATH; return 0; } + +/** cifs_dfs_query_info_nonascii_quirk + * Handle weird Windows SMB server behaviour. It responds with + * STATUS_OBJECT_NAME_INVALID code to SMB2 QUERY_INFO request + * for "\<server>\<dfsname>\<linkpath>" DFS reference, + * where <dfsname> contains non-ASCII unicode symbols. + * + * Check such DFS reference and emulate -ENOENT if it is actual. + */ +int cifs_dfs_query_info_nonascii_quirk(const unsigned int xid, + struct cifs_tcon *tcon, + struct cifs_sb_info *cifs_sb, + const char *linkpath) +{ + char *treename, *dfspath, sep; + int treenamelen, linkpathlen, rc; + + treename = tcon->treeName; + /* MS-DFSC: All paths in REQ_GET_DFS_REFERRAL and RESP_GET_DFS_REFERRAL + * messages MUST be encoded with exactly one leading backslash, not two + * leading backslashes. + */ + sep = CIFS_DIR_SEP(cifs_sb); + if (treename[0] == sep && treename[1] == sep) + treename++; + linkpathlen = strlen(linkpath); + treenamelen = strnlen(treename, MAX_TREE_SIZE + 1); + dfspath = kzalloc(treenamelen + linkpathlen + 1, GFP_KERNEL); + if (!dfspath) + return -ENOMEM; + if (treenamelen) + memcpy(dfspath, treename, treenamelen); + memcpy(dfspath + treenamelen, linkpath, linkpathlen); + rc = dfs_cache_find(xid, tcon->ses, cifs_sb->local_nls, + cifs_remap(cifs_sb), dfspath, NULL, NULL); + if (rc == 0) { + cifs_dbg(FYI, "DFS ref '%s' is found, emulate -EREMOTE\n", + dfspath); + rc = -EREMOTE; + } else if (rc == -EEXIST) { + cifs_dbg(FYI, "DFS ref '%s' is not found, emulate -ENOENT\n", + dfspath); + rc = -ENOENT; + } else { + cifs_dbg(FYI, "%s: dfs_cache_find returned %d\n", __func__, rc); + } + kfree(dfspath); + return rc; +} #endif diff --git a/fs/cifs/netmisc.c b/fs/cifs/netmisc.c index fa9fbd6a819c..ebe236b9d9f5 100644 --- a/fs/cifs/netmisc.c +++ b/fs/cifs/netmisc.c @@ -896,10 +896,7 @@ map_and_check_smb_error(struct mid_q_entry *mid, bool logErr) if (class == ERRSRV && code == ERRbaduid) { cifs_dbg(FYI, "Server returned 0x%x, reconnecting session...\n", code); - spin_lock(&GlobalMid_Lock); - if (mid->server->tcpStatus != CifsExiting) - mid->server->tcpStatus = CifsNeedReconnect; - spin_unlock(&GlobalMid_Lock); + cifs_reconnect(mid->server, false); } } diff --git a/fs/cifs/ntlmssp.h b/fs/cifs/ntlmssp.h index fe707f45da89..298458404252 100644 --- a/fs/cifs/ntlmssp.h +++ b/fs/cifs/ntlmssp.h @@ -40,7 +40,7 @@ #define NTLMSSP_REQUEST_NON_NT_KEY 0x400000 #define NTLMSSP_NEGOTIATE_TARGET_INFO 0x800000 /* #define reserved4 0x1000000 */ -#define NTLMSSP_NEGOTIATE_VERSION 0x2000000 /* we do not set */ +#define NTLMSSP_NEGOTIATE_VERSION 0x2000000 /* we only set for SMB2+ */ /* #define reserved3 0x4000000 */ /* #define reserved2 0x8000000 */ /* #define reserved1 0x10000000 */ @@ -87,6 +87,30 @@ typedef struct _NEGOTIATE_MESSAGE { /* followed by WorkstationString */ } __attribute__((packed)) NEGOTIATE_MESSAGE, *PNEGOTIATE_MESSAGE; +#define NTLMSSP_REVISION_W2K3 0x0F + +/* See MS-NLMP section 2.2.2.10 */ +struct ntlmssp_version { + __u8 ProductMajorVersion; + __u8 ProductMinorVersion; + __le16 ProductBuild; /* we send the cifs.ko module version here */ + __u8 Reserved[3]; + __u8 NTLMRevisionCurrent; /* currently 0x0F */ +} __packed; + +/* see MS-NLMP section 2.2.1.1 */ +struct negotiate_message { + __u8 Signature[sizeof(NTLMSSP_SIGNATURE)]; + __le32 MessageType; /* NtLmNegotiate = 1 */ + __le32 NegotiateFlags; + SECURITY_BUFFER DomainName; /* RFC 1001 style and ASCII */ + SECURITY_BUFFER WorkstationName; /* RFC 1001 and ASCII */ + struct ntlmssp_version Version; + /* SECURITY_BUFFER */ + char DomainString[0]; + /* followed by WorkstationString */ +} __packed; + typedef struct _CHALLENGE_MESSAGE { __u8 Signature[sizeof(NTLMSSP_SIGNATURE)]; __le32 MessageType; /* NtLmChallenge = 2 */ @@ -121,7 +145,13 @@ typedef struct _AUTHENTICATE_MESSAGE { int decode_ntlmssp_challenge(char *bcc_ptr, int blob_len, struct cifs_ses *ses); int build_ntlmssp_negotiate_blob(unsigned char **pbuffer, u16 *buflen, struct cifs_ses *ses, + struct TCP_Server_Info *server, + const struct nls_table *nls_cp); +int build_ntlmssp_smb3_negotiate_blob(unsigned char **pbuffer, u16 *buflen, + struct cifs_ses *ses, + struct TCP_Server_Info *server, const struct nls_table *nls_cp); int build_ntlmssp_auth_blob(unsigned char **pbuffer, u16 *buflen, struct cifs_ses *ses, + struct TCP_Server_Info *server, const struct nls_table *nls_cp); diff --git a/fs/cifs/sess.c b/fs/cifs/sess.c index 035dc3e245dc..dc3b16d1be09 100644 --- a/fs/cifs/sess.c +++ b/fs/cifs/sess.c @@ -17,6 +17,8 @@ #include "nterr.h" #include <linux/utsname.h> #include <linux/slab.h> +#include <linux/version.h> +#include "cifsfs.h" #include "cifs_spnego.h" #include "smb2proto.h" #include "fs_context.h" @@ -65,6 +67,55 @@ bool is_ses_using_iface(struct cifs_ses *ses, struct cifs_server_iface *iface) return false; } +/* channel helper functions. assumed that chan_lock is held by caller. */ + +unsigned int +cifs_ses_get_chan_index(struct cifs_ses *ses, + struct TCP_Server_Info *server) +{ + unsigned int i; + + for (i = 0; i < ses->chan_count; i++) { + if (ses->chans[i].server == server) + return i; + } + + /* If we didn't find the channel, it is likely a bug */ + WARN_ON(1); + return 0; +} + +void +cifs_chan_set_need_reconnect(struct cifs_ses *ses, + struct TCP_Server_Info *server) +{ + unsigned int chan_index = cifs_ses_get_chan_index(ses, server); + + set_bit(chan_index, &ses->chans_need_reconnect); + cifs_dbg(FYI, "Set reconnect bitmask for chan %u; now 0x%lx\n", + chan_index, ses->chans_need_reconnect); +} + +void +cifs_chan_clear_need_reconnect(struct cifs_ses *ses, + struct TCP_Server_Info *server) +{ + unsigned int chan_index = cifs_ses_get_chan_index(ses, server); + + clear_bit(chan_index, &ses->chans_need_reconnect); + cifs_dbg(FYI, "Cleared reconnect bitmask for chan %u; now 0x%lx\n", + chan_index, ses->chans_need_reconnect); +} + +bool +cifs_chan_needs_reconnect(struct cifs_ses *ses, + struct TCP_Server_Info *server) +{ + unsigned int chan_index = cifs_ses_get_chan_index(ses, server); + + return CIFS_CHAN_NEEDS_RECONNECT(ses, chan_index); +} + /* returns number of channels added */ int cifs_try_adding_channels(struct cifs_sb_info *cifs_sb, struct cifs_ses *ses) { @@ -87,10 +138,10 @@ int cifs_try_adding_channels(struct cifs_sb_info *cifs_sb, struct cifs_ses *ses) left = ses->chan_max - ses->chan_count; if (left <= 0) { + spin_unlock(&ses->chan_lock); cifs_dbg(FYI, "ses already at max_channels (%zu), nothing to open\n", ses->chan_max); - spin_unlock(&ses->chan_lock); return 0; } @@ -261,9 +312,8 @@ cifs_ses_add_channel(struct cifs_sb_info *cifs_sb, struct cifs_ses *ses, chan_server = cifs_get_tcp_session(&ctx, ses->server); - mutex_lock(&ses->session_mutex); spin_lock(&ses->chan_lock); - chan = ses->binding_chan = &ses->chans[ses->chan_count]; + chan = &ses->chans[ses->chan_count]; chan->server = chan_server; if (IS_ERR(chan->server)) { rc = PTR_ERR(chan->server); @@ -271,8 +321,15 @@ cifs_ses_add_channel(struct cifs_sb_info *cifs_sb, struct cifs_ses *ses, spin_unlock(&ses->chan_lock); goto out; } + ses->chan_count++; + atomic_set(&ses->chan_seq, 0); + + /* Mark this channel as needing connect/setup */ + cifs_chan_set_need_reconnect(ses, chan->server); + spin_unlock(&ses->chan_lock); + mutex_lock(&ses->session_mutex); /* * We need to allocate the server crypto now as we will need * to sign packets before we generate the channel signing key @@ -281,37 +338,29 @@ cifs_ses_add_channel(struct cifs_sb_info *cifs_sb, struct cifs_ses *ses, rc = smb311_crypto_shash_allocate(chan->server); if (rc) { cifs_dbg(VFS, "%s: crypto alloc failed\n", __func__); + mutex_unlock(&ses->session_mutex); goto out; } - ses->binding = true; - rc = cifs_negotiate_protocol(xid, ses); - if (rc) - goto out; - - rc = cifs_setup_session(xid, ses, cifs_sb->local_nls); - if (rc) - goto out; - - /* success, put it on the list - * XXX: sharing ses between 2 tcp servers is not possible, the - * way "internal" linked lists works in linux makes element - * only able to belong to one list - * - * the binding session is already established so the rest of - * the code should be able to look it up, no need to add the - * ses to the new server. - */ + rc = cifs_negotiate_protocol(xid, ses, chan->server); + if (!rc) + rc = cifs_setup_session(xid, ses, chan->server, cifs_sb->local_nls); - spin_lock(&ses->chan_lock); - ses->chan_count++; - atomic_set(&ses->chan_seq, 0); - spin_unlock(&ses->chan_lock); + mutex_unlock(&ses->session_mutex); out: - ses->binding = false; - ses->binding_chan = NULL; - mutex_unlock(&ses->session_mutex); + if (rc && chan->server) { + spin_lock(&ses->chan_lock); + /* we rely on all bits beyond chan_count to be clear */ + cifs_chan_clear_need_reconnect(ses, chan->server); + ses->chan_count--; + /* + * chan_count should never reach 0 as at least the primary + * channel is always allocated + */ + WARN_ON(ses->chan_count < 1); + spin_unlock(&ses->chan_lock); + } if (rc && chan->server) cifs_put_tcp_session(chan->server, 0); @@ -319,20 +368,9 @@ out: return rc; } -/* Mark all session channels for reconnect */ -void cifs_ses_mark_for_reconnect(struct cifs_ses *ses) -{ - int i; - - for (i = 0; i < ses->chan_count; i++) { - spin_lock(&GlobalMid_Lock); - if (ses->chans[i].server->tcpStatus != CifsExiting) - ses->chans[i].server->tcpStatus = CifsNeedReconnect; - spin_unlock(&GlobalMid_Lock); - } -} - -static __u32 cifs_ssetup_hdr(struct cifs_ses *ses, SESSION_SETUP_ANDX *pSMB) +static __u32 cifs_ssetup_hdr(struct cifs_ses *ses, + struct TCP_Server_Info *server, + SESSION_SETUP_ANDX *pSMB) { __u32 capabilities = 0; @@ -345,7 +383,7 @@ static __u32 cifs_ssetup_hdr(struct cifs_ses *ses, SESSION_SETUP_ANDX *pSMB) pSMB->req.MaxBufferSize = cpu_to_le16(min_t(u32, CIFSMaxBufSize + MAX_CIFS_HDR_SIZE - 4, USHRT_MAX)); - pSMB->req.MaxMpxCount = cpu_to_le16(ses->server->maxReq); + pSMB->req.MaxMpxCount = cpu_to_le16(server->maxReq); pSMB->req.VcNumber = cpu_to_le16(1); /* Now no need to set SMBFLG_CASELESS or obsolete CANONICAL PATH */ @@ -356,7 +394,7 @@ static __u32 cifs_ssetup_hdr(struct cifs_ses *ses, SESSION_SETUP_ANDX *pSMB) capabilities = CAP_LARGE_FILES | CAP_NT_SMBS | CAP_LEVEL_II_OPLOCKS | CAP_LARGE_WRITE_X | CAP_LARGE_READ_X; - if (ses->server->sign) + if (server->sign) pSMB->req.hdr.Flags2 |= SMBFLG2_SECURITY_SIGNATURE; if (ses->capabilities & CAP_UNICODE) { @@ -719,10 +757,10 @@ static inline void cifs_security_buffer_from_str(SECURITY_BUFFER *pbuf, int build_ntlmssp_negotiate_blob(unsigned char **pbuffer, u16 *buflen, struct cifs_ses *ses, + struct TCP_Server_Info *server, const struct nls_table *nls_cp) { int rc = 0; - struct TCP_Server_Info *server = cifs_ses_server(ses); NEGOTIATE_MESSAGE *sec_blob; __u32 flags; unsigned char *tmp; @@ -773,9 +811,78 @@ setup_ntlm_neg_ret: return rc; } +/* + * Build ntlmssp blob with additional fields, such as version, + * supported by modern servers. For safety limit to SMB3 or later + * See notes in MS-NLMP Section 2.2.2.1 e.g. + */ +int build_ntlmssp_smb3_negotiate_blob(unsigned char **pbuffer, + u16 *buflen, + struct cifs_ses *ses, + struct TCP_Server_Info *server, + const struct nls_table *nls_cp) +{ + int rc = 0; + struct negotiate_message *sec_blob; + __u32 flags; + unsigned char *tmp; + int len; + + len = size_of_ntlmssp_blob(ses, sizeof(struct negotiate_message)); + *pbuffer = kmalloc(len, GFP_KERNEL); + if (!*pbuffer) { + rc = -ENOMEM; + cifs_dbg(VFS, "Error %d during NTLMSSP allocation\n", rc); + *buflen = 0; + goto setup_ntlm_smb3_neg_ret; + } + sec_blob = (struct negotiate_message *)*pbuffer; + + memset(*pbuffer, 0, sizeof(struct negotiate_message)); + memcpy(sec_blob->Signature, NTLMSSP_SIGNATURE, 8); + sec_blob->MessageType = NtLmNegotiate; + + /* BB is NTLMV2 session security format easier to use here? */ + flags = NTLMSSP_NEGOTIATE_56 | NTLMSSP_REQUEST_TARGET | + NTLMSSP_NEGOTIATE_128 | NTLMSSP_NEGOTIATE_UNICODE | + NTLMSSP_NEGOTIATE_NTLM | NTLMSSP_NEGOTIATE_EXTENDED_SEC | + NTLMSSP_NEGOTIATE_ALWAYS_SIGN | NTLMSSP_NEGOTIATE_SEAL | + NTLMSSP_NEGOTIATE_SIGN | NTLMSSP_NEGOTIATE_VERSION; + if (!server->session_estab || ses->ntlmssp->sesskey_per_smbsess) + flags |= NTLMSSP_NEGOTIATE_KEY_XCH; + + sec_blob->Version.ProductMajorVersion = LINUX_VERSION_MAJOR; + sec_blob->Version.ProductMinorVersion = LINUX_VERSION_PATCHLEVEL; + sec_blob->Version.ProductBuild = cpu_to_le16(SMB3_PRODUCT_BUILD); + sec_blob->Version.NTLMRevisionCurrent = NTLMSSP_REVISION_W2K3; + + tmp = *pbuffer + sizeof(struct negotiate_message); + ses->ntlmssp->client_flags = flags; + sec_blob->NegotiateFlags = cpu_to_le32(flags); + + /* these fields should be null in negotiate phase MS-NLMP 3.1.5.1.1 */ + cifs_security_buffer_from_str(&sec_blob->DomainName, + NULL, + CIFS_MAX_DOMAINNAME_LEN, + *pbuffer, &tmp, + nls_cp); + + cifs_security_buffer_from_str(&sec_blob->WorkstationName, + NULL, + CIFS_MAX_WORKSTATION_LEN, + *pbuffer, &tmp, + nls_cp); + + *buflen = tmp - *pbuffer; +setup_ntlm_smb3_neg_ret: + return rc; +} + + int build_ntlmssp_auth_blob(unsigned char **pbuffer, u16 *buflen, struct cifs_ses *ses, + struct TCP_Server_Info *server, const struct nls_table *nls_cp) { int rc; @@ -912,6 +1019,7 @@ cifs_select_sectype(struct TCP_Server_Info *server, enum securityEnum requested) struct sess_data { unsigned int xid; struct cifs_ses *ses; + struct TCP_Server_Info *server; struct nls_table *nls_cp; void (*func)(struct sess_data *); int result; @@ -978,31 +1086,27 @@ static int sess_establish_session(struct sess_data *sess_data) { struct cifs_ses *ses = sess_data->ses; + struct TCP_Server_Info *server = sess_data->server; - mutex_lock(&ses->server->srv_mutex); - if (!ses->server->session_estab) { - if (ses->server->sign) { - ses->server->session_key.response = + mutex_lock(&server->srv_mutex); + if (!server->session_estab) { + if (server->sign) { + server->session_key.response = kmemdup(ses->auth_key.response, ses->auth_key.len, GFP_KERNEL); - if (!ses->server->session_key.response) { - mutex_unlock(&ses->server->srv_mutex); + if (!server->session_key.response) { + mutex_unlock(&server->srv_mutex); return -ENOMEM; } - ses->server->session_key.len = + server->session_key.len = ses->auth_key.len; } - ses->server->sequence_number = 0x2; - ses->server->session_estab = true; + server->sequence_number = 0x2; + server->session_estab = true; } - mutex_unlock(&ses->server->srv_mutex); + mutex_unlock(&server->srv_mutex); cifs_dbg(FYI, "CIFS session established successfully\n"); - spin_lock(&GlobalMid_Lock); - ses->status = CifsGood; - ses->need_reconnect = false; - spin_unlock(&GlobalMid_Lock); - return 0; } @@ -1036,6 +1140,7 @@ sess_auth_ntlmv2(struct sess_data *sess_data) SESSION_SETUP_ANDX *pSMB; char *bcc_ptr; struct cifs_ses *ses = sess_data->ses; + struct TCP_Server_Info *server = sess_data->server; __u32 capabilities; __u16 bytes_remaining; @@ -1047,7 +1152,7 @@ sess_auth_ntlmv2(struct sess_data *sess_data) pSMB = (SESSION_SETUP_ANDX *)sess_data->iov[0].iov_base; bcc_ptr = sess_data->iov[2].iov_base; - capabilities = cifs_ssetup_hdr(ses, pSMB); + capabilities = cifs_ssetup_hdr(ses, server, pSMB); pSMB->req_no_secext.Capabilities = cpu_to_le32(capabilities); @@ -1145,6 +1250,7 @@ sess_auth_kerberos(struct sess_data *sess_data) SESSION_SETUP_ANDX *pSMB; char *bcc_ptr; struct cifs_ses *ses = sess_data->ses; + struct TCP_Server_Info *server = sess_data->server; __u32 capabilities; __u16 bytes_remaining; struct key *spnego_key = NULL; @@ -1159,9 +1265,9 @@ sess_auth_kerberos(struct sess_data *sess_data) pSMB = (SESSION_SETUP_ANDX *)sess_data->iov[0].iov_base; bcc_ptr = sess_data->iov[2].iov_base; - capabilities = cifs_ssetup_hdr(ses, pSMB); + capabilities = cifs_ssetup_hdr(ses, server, pSMB); - spnego_key = cifs_get_spnego_key(ses); + spnego_key = cifs_get_spnego_key(ses, server); if (IS_ERR(spnego_key)) { rc = PTR_ERR(spnego_key); spnego_key = NULL; @@ -1285,12 +1391,13 @@ _sess_auth_rawntlmssp_assemble_req(struct sess_data *sess_data) { SESSION_SETUP_ANDX *pSMB; struct cifs_ses *ses = sess_data->ses; + struct TCP_Server_Info *server = sess_data->server; __u32 capabilities; char *bcc_ptr; pSMB = (SESSION_SETUP_ANDX *)sess_data->iov[0].iov_base; - capabilities = cifs_ssetup_hdr(ses, pSMB); + capabilities = cifs_ssetup_hdr(ses, server, pSMB); if ((pSMB->req.hdr.Flags2 & SMBFLG2_UNICODE) == 0) { cifs_dbg(VFS, "NTLMSSP requires Unicode support\n"); return -ENOSYS; @@ -1324,6 +1431,7 @@ sess_auth_rawntlmssp_negotiate(struct sess_data *sess_data) struct smb_hdr *smb_buf; SESSION_SETUP_ANDX *pSMB; struct cifs_ses *ses = sess_data->ses; + struct TCP_Server_Info *server = sess_data->server; __u16 bytes_remaining; char *bcc_ptr; unsigned char *ntlmsspblob = NULL; @@ -1351,10 +1459,10 @@ sess_auth_rawntlmssp_negotiate(struct sess_data *sess_data) /* Build security blob before we assemble the request */ rc = build_ntlmssp_negotiate_blob(&ntlmsspblob, - &blob_len, ses, + &blob_len, ses, server, sess_data->nls_cp); if (rc) - goto out; + goto out_free_ntlmsspblob; sess_data->iov[1].iov_len = blob_len; sess_data->iov[1].iov_base = ntlmsspblob; @@ -1362,7 +1470,7 @@ sess_auth_rawntlmssp_negotiate(struct sess_data *sess_data) rc = _sess_auth_rawntlmssp_assemble_req(sess_data); if (rc) - goto out; + goto out_free_ntlmsspblob; rc = sess_sendreceive(sess_data); @@ -1376,14 +1484,14 @@ sess_auth_rawntlmssp_negotiate(struct sess_data *sess_data) rc = 0; if (rc) - goto out; + goto out_free_ntlmsspblob; cifs_dbg(FYI, "rawntlmssp session setup challenge phase\n"); if (smb_buf->WordCount != 4) { rc = -EIO; cifs_dbg(VFS, "bad word count %d\n", smb_buf->WordCount); - goto out; + goto out_free_ntlmsspblob; } ses->Suid = smb_buf->Uid; /* UID left in wire format (le) */ @@ -1397,10 +1505,13 @@ sess_auth_rawntlmssp_negotiate(struct sess_data *sess_data) cifs_dbg(VFS, "bad security blob length %d\n", blob_len); rc = -EINVAL; - goto out; + goto out_free_ntlmsspblob; } rc = decode_ntlmssp_challenge(bcc_ptr, blob_len, ses); + +out_free_ntlmsspblob: + kfree(ntlmsspblob); out: sess_free_buffer(sess_data); @@ -1426,6 +1537,7 @@ sess_auth_rawntlmssp_authenticate(struct sess_data *sess_data) struct smb_hdr *smb_buf; SESSION_SETUP_ANDX *pSMB; struct cifs_ses *ses = sess_data->ses; + struct TCP_Server_Info *server = sess_data->server; __u16 bytes_remaining; char *bcc_ptr; unsigned char *ntlmsspblob = NULL; @@ -1442,7 +1554,8 @@ sess_auth_rawntlmssp_authenticate(struct sess_data *sess_data) pSMB = (SESSION_SETUP_ANDX *)sess_data->iov[0].iov_base; smb_buf = (struct smb_hdr *)pSMB; rc = build_ntlmssp_auth_blob(&ntlmsspblob, - &blob_len, ses, sess_data->nls_cp); + &blob_len, ses, server, + sess_data->nls_cp); if (rc) goto out_free_ntlmsspblob; sess_data->iov[1].iov_len = blob_len; @@ -1513,7 +1626,7 @@ out_free_ntlmsspblob: out: sess_free_buffer(sess_data); - if (!rc) + if (!rc) rc = sess_establish_session(sess_data); /* Cleanup */ @@ -1526,11 +1639,13 @@ out: sess_data->result = rc; } -static int select_sec(struct cifs_ses *ses, struct sess_data *sess_data) +static int select_sec(struct sess_data *sess_data) { int type; + struct cifs_ses *ses = sess_data->ses; + struct TCP_Server_Info *server = sess_data->server; - type = cifs_select_sectype(ses->server, ses->sectype); + type = cifs_select_sectype(server, ses->sectype); cifs_dbg(FYI, "sess setup type %d\n", type); if (type == Unspecified) { cifs_dbg(VFS, "Unable to select appropriate authentication method!\n"); @@ -1561,7 +1676,8 @@ static int select_sec(struct cifs_ses *ses, struct sess_data *sess_data) } int CIFS_SessSetup(const unsigned int xid, struct cifs_ses *ses, - const struct nls_table *nls_cp) + struct TCP_Server_Info *server, + const struct nls_table *nls_cp) { int rc = 0; struct sess_data *sess_data; @@ -1575,15 +1691,16 @@ int CIFS_SessSetup(const unsigned int xid, struct cifs_ses *ses, if (!sess_data) return -ENOMEM; - rc = select_sec(ses, sess_data); - if (rc) - goto out; - sess_data->xid = xid; sess_data->ses = ses; + sess_data->server = server; sess_data->buf0_type = CIFS_NO_BUFFER; sess_data->nls_cp = (struct nls_table *) nls_cp; + rc = select_sec(sess_data); + if (rc) + goto out; + while (sess_data->func) sess_data->func(sess_data); diff --git a/fs/cifs/smb1ops.c b/fs/cifs/smb1ops.c index 3b83839fc2c2..8272c91e15ef 100644 --- a/fs/cifs/smb1ops.c +++ b/fs/cifs/smb1ops.c @@ -7,6 +7,7 @@ #include <linux/pagemap.h> #include <linux/vfs.h> +#include <uapi/linux/magic.h> #include "cifsglob.h" #include "cifsproto.h" #include "cifs_debug.h" @@ -163,7 +164,7 @@ cifs_get_next_mid(struct TCP_Server_Info *server) { __u64 mid = 0; __u16 last_mid, cur_mid; - bool collision; + bool collision, reconnect = false; spin_lock(&GlobalMid_Lock); @@ -215,7 +216,7 @@ cifs_get_next_mid(struct TCP_Server_Info *server) * an eventual reconnect to clean out the pending_mid_q. */ if (num_mids > 32768) - server->tcpStatus = CifsNeedReconnect; + reconnect = true; if (!collision) { mid = (__u64)cur_mid; @@ -225,6 +226,13 @@ cifs_get_next_mid(struct TCP_Server_Info *server) cur_mid++; } spin_unlock(&GlobalMid_Lock); + + if (reconnect) { + spin_lock(&cifs_tcp_ses_lock); + server->tcpStatus = CifsNeedReconnect; + spin_unlock(&cifs_tcp_ses_lock); + } + return mid; } @@ -414,14 +422,16 @@ cifs_need_neg(struct TCP_Server_Info *server) } static int -cifs_negotiate(const unsigned int xid, struct cifs_ses *ses) +cifs_negotiate(const unsigned int xid, + struct cifs_ses *ses, + struct TCP_Server_Info *server) { int rc; - rc = CIFSSMBNegotiate(xid, ses); + rc = CIFSSMBNegotiate(xid, ses, server); if (rc == -EAGAIN) { /* retry only once on 1st time connection */ - set_credits(ses->server, 1); - rc = CIFSSMBNegotiate(xid, ses); + set_credits(server, 1); + rc = CIFSSMBNegotiate(xid, ses, server); if (rc == -EAGAIN) rc = -EHOSTDOWN; } @@ -878,7 +888,7 @@ cifs_queryfs(const unsigned int xid, struct cifs_tcon *tcon, { int rc = -EOPNOTSUPP; - buf->f_type = CIFS_MAGIC_NUMBER; + buf->f_type = CIFS_SUPER_MAGIC; /* * We could add a second check for a QFS Unix capability bit diff --git a/fs/cifs/smb2glob.h b/fs/cifs/smb2glob.h index ca692b2283cd..4125fd113cfb 100644 --- a/fs/cifs/smb2glob.h +++ b/fs/cifs/smb2glob.h @@ -13,8 +13,6 @@ #ifndef _SMB2_GLOB_H #define _SMB2_GLOB_H -#define SMB2_MAGIC_NUMBER 0xFE534D42 - /* ***************************************************************** * Constants go here diff --git a/fs/cifs/smb2misc.c b/fs/cifs/smb2misc.c index cdcdef32759e..b25623e3fe3d 100644 --- a/fs/cifs/smb2misc.c +++ b/fs/cifs/smb2misc.c @@ -847,16 +847,17 @@ smb2_handle_cancelled_mid(struct mid_q_entry *mid, struct TCP_Server_Info *serve * SMB2 header. * * @ses: server session structure + * @server: pointer to server info * @iov: array containing the SMB request we will send to the server * @nvec: number of array entries for the iov */ int -smb311_update_preauth_hash(struct cifs_ses *ses, struct kvec *iov, int nvec) +smb311_update_preauth_hash(struct cifs_ses *ses, struct TCP_Server_Info *server, + struct kvec *iov, int nvec) { int i, rc; struct sdesc *d; struct smb2_hdr *hdr; - struct TCP_Server_Info *server = cifs_ses_server(ses); hdr = (struct smb2_hdr *)iov[0].iov_base; /* neg prot are always taken */ diff --git a/fs/cifs/smb2ops.c b/fs/cifs/smb2ops.c index c5b1dea54ebc..af5d0830bc8a 100644 --- a/fs/cifs/smb2ops.c +++ b/fs/cifs/smb2ops.c @@ -13,6 +13,7 @@ #include <linux/sort.h> #include <crypto/aead.h> #include <linux/fiemap.h> +#include <uapi/linux/magic.h> #include "cifsfs.h" #include "cifsglob.h" #include "smb2pdu.h" @@ -121,9 +122,13 @@ smb2_add_credits(struct TCP_Server_Info *server, optype, scredits, add); } + spin_lock(&cifs_tcp_ses_lock); if (server->tcpStatus == CifsNeedReconnect - || server->tcpStatus == CifsExiting) + || server->tcpStatus == CifsExiting) { + spin_unlock(&cifs_tcp_ses_lock); return; + } + spin_unlock(&cifs_tcp_ses_lock); switch (rc) { case -1: @@ -208,11 +213,15 @@ smb2_wait_mtu_credits(struct TCP_Server_Info *server, unsigned int size, return rc; spin_lock(&server->req_lock); } else { + spin_unlock(&server->req_lock); + spin_lock(&cifs_tcp_ses_lock); if (server->tcpStatus == CifsExiting) { - spin_unlock(&server->req_lock); + spin_unlock(&cifs_tcp_ses_lock); return -ENOENT; } + spin_unlock(&cifs_tcp_ses_lock); + spin_lock(&server->req_lock); scredits = server->credits; /* can deadlock with reopen */ if (scredits <= 8) { @@ -384,14 +393,16 @@ smb2_need_neg(struct TCP_Server_Info *server) } static int -smb2_negotiate(const unsigned int xid, struct cifs_ses *ses) +smb2_negotiate(const unsigned int xid, + struct cifs_ses *ses, + struct TCP_Server_Info *server) { int rc; spin_lock(&GlobalMid_Lock); - cifs_ses_server(ses)->CurrentMid = 0; + server->CurrentMid = 0; spin_unlock(&GlobalMid_Lock); - rc = SMB2_negotiate(xid, ses); + rc = SMB2_negotiate(xid, ses, server); /* BB we probably don't need to retry with modern servers */ if (rc == -EAGAIN) rc = -EHOSTDOWN; @@ -2747,7 +2758,7 @@ smb2_queryfs(const unsigned int xid, struct cifs_tcon *tcon, goto qfs_exit; rsp = (struct smb2_query_info_rsp *)rsp_iov.iov_base; - buf->f_type = SMB2_MAGIC_NUMBER; + buf->f_type = SMB2_SUPER_MAGIC; info = (struct smb2_fs_full_size_info *)( le16_to_cpu(rsp->OutputBufferOffset) + (char *)rsp); rc = smb2_validate_iov(le16_to_cpu(rsp->OutputBufferOffset), @@ -2789,7 +2800,7 @@ smb311_queryfs(const unsigned int xid, struct cifs_tcon *tcon, rc = SMB311_posix_qfs_info(xid, tcon, fid.persistent_fid, fid.volatile_fid, buf); - buf->f_type = SMB2_MAGIC_NUMBER; + buf->f_type = SMB2_SUPER_MAGIC; SMB2_close(xid, tcon, fid.persistent_fid, fid.volatile_fid); return rc; } @@ -4808,7 +4819,7 @@ handle_read_data(struct TCP_Server_Info *server, struct mid_q_entry *mid, if (server->ops->is_session_expired && server->ops->is_session_expired(buf)) { if (!is_offloaded) - cifs_reconnect(server); + cifs_reconnect(server, true); return -1; } @@ -4981,10 +4992,12 @@ static void smb2_decrypt_offload(struct work_struct *work) mid->callback(mid); } else { + spin_lock(&cifs_tcp_ses_lock); spin_lock(&GlobalMid_Lock); if (dw->server->tcpStatus == CifsNeedReconnect) { mid->mid_state = MID_RETRY_NEEDED; spin_unlock(&GlobalMid_Lock); + spin_unlock(&cifs_tcp_ses_lock); mid->callback(mid); } else { mid->mid_state = MID_REQUEST_SUBMITTED; @@ -4992,6 +5005,7 @@ static void smb2_decrypt_offload(struct work_struct *work) list_add_tail(&mid->qhead, &dw->server->pending_mid_q); spin_unlock(&GlobalMid_Lock); + spin_unlock(&cifs_tcp_ses_lock); } } cifs_mid_q_entry_release(mid); @@ -5221,13 +5235,13 @@ smb3_receive_transform(struct TCP_Server_Info *server, sizeof(struct smb2_hdr)) { cifs_server_dbg(VFS, "Transform message is too small (%u)\n", pdu_length); - cifs_reconnect(server); + cifs_reconnect(server, true); return -ECONNABORTED; } if (pdu_length < orig_len + sizeof(struct smb2_transform_hdr)) { cifs_server_dbg(VFS, "Transform message is broken\n"); - cifs_reconnect(server); + cifs_reconnect(server, true); return -ECONNABORTED; } diff --git a/fs/cifs/smb2pdu.c b/fs/cifs/smb2pdu.c index 8b3670388cda..7e7909b1ae11 100644 --- a/fs/cifs/smb2pdu.c +++ b/fs/cifs/smb2pdu.c @@ -162,6 +162,7 @@ smb2_reconnect(__le16 smb2_command, struct cifs_tcon *tcon, if (smb2_command == SMB2_TREE_CONNECT || smb2_command == SMB2_IOCTL) return 0; + spin_lock(&cifs_tcp_ses_lock); if (tcon->tidStatus == CifsExiting) { /* * only tree disconnect, open, and write, @@ -171,11 +172,13 @@ smb2_reconnect(__le16 smb2_command, struct cifs_tcon *tcon, if ((smb2_command != SMB2_WRITE) && (smb2_command != SMB2_CREATE) && (smb2_command != SMB2_TREE_DISCONNECT)) { + spin_unlock(&cifs_tcp_ses_lock); cifs_dbg(FYI, "can not send cmd %d while umounting\n", smb2_command); return -ENODEV; } } + spin_unlock(&cifs_tcp_ses_lock); if ((!tcon->ses) || (tcon->ses->status == CifsExiting) || (!tcon->ses->server) || !server) return -EIO; @@ -214,8 +217,12 @@ smb2_reconnect(__le16 smb2_command, struct cifs_tcon *tcon, } /* are we still trying to reconnect? */ - if (server->tcpStatus != CifsNeedReconnect) + spin_lock(&cifs_tcp_ses_lock); + if (server->tcpStatus != CifsNeedReconnect) { + spin_unlock(&cifs_tcp_ses_lock); break; + } + spin_unlock(&cifs_tcp_ses_lock); if (retries && --retries) continue; @@ -232,64 +239,74 @@ smb2_reconnect(__le16 smb2_command, struct cifs_tcon *tcon, retries = server->nr_targets; } - if (!tcon->ses->need_reconnect && !tcon->need_reconnect) + spin_lock(&ses->chan_lock); + if (!cifs_chan_needs_reconnect(ses, server) && !tcon->need_reconnect) { + spin_unlock(&ses->chan_lock); return 0; + } + spin_unlock(&ses->chan_lock); + cifs_dbg(FYI, "sess reconnect mask: 0x%lx, tcon reconnect: %d", + tcon->ses->chans_need_reconnect, + tcon->need_reconnect); nls_codepage = load_nls_default(); /* - * need to prevent multiple threads trying to simultaneously reconnect - * the same SMB session - */ - mutex_lock(&tcon->ses->session_mutex); - - /* * Recheck after acquire mutex. If another thread is negotiating * and the server never sends an answer the socket will be closed * and tcpStatus set to reconnect. */ + spin_lock(&cifs_tcp_ses_lock); if (server->tcpStatus == CifsNeedReconnect) { + spin_unlock(&cifs_tcp_ses_lock); rc = -EHOSTDOWN; - mutex_unlock(&tcon->ses->session_mutex); goto out; } + spin_unlock(&cifs_tcp_ses_lock); /* - * If we are reconnecting an extra channel, bind + * need to prevent multiple threads trying to simultaneously + * reconnect the same SMB session */ - if (CIFS_SERVER_IS_CHAN(server)) { - ses->binding = true; - ses->binding_chan = cifs_ses_find_chan(ses, server); + spin_lock(&ses->chan_lock); + if (!cifs_chan_needs_reconnect(ses, server)) { + spin_unlock(&ses->chan_lock); + + /* this means that we only need to tree connect */ + if (tcon->need_reconnect) + goto skip_sess_setup; + + goto out; } + spin_unlock(&ses->chan_lock); - rc = cifs_negotiate_protocol(0, tcon->ses); - if (!rc && tcon->ses->need_reconnect) { - rc = cifs_setup_session(0, tcon->ses, nls_codepage); + mutex_lock(&ses->session_mutex); + rc = cifs_negotiate_protocol(0, ses, server); + if (!rc) { + rc = cifs_setup_session(0, ses, server, nls_codepage); if ((rc == -EACCES) && !tcon->retry) { + mutex_unlock(&ses->session_mutex); rc = -EHOSTDOWN; - ses->binding = false; - ses->binding_chan = NULL; - mutex_unlock(&tcon->ses->session_mutex); goto failed; } + } else { + mutex_unlock(&ses->session_mutex); + goto out; } - /* - * End of channel binding - */ - ses->binding = false; - ses->binding_chan = NULL; + mutex_unlock(&ses->session_mutex); - if (rc || !tcon->need_reconnect) { - mutex_unlock(&tcon->ses->session_mutex); +skip_sess_setup: + mutex_lock(&ses->session_mutex); + if (!tcon->need_reconnect) { + mutex_unlock(&ses->session_mutex); goto out; } - cifs_mark_open_files_invalid(tcon); if (tcon->use_persistent) tcon->need_reopen_files = true; rc = cifs_tree_connect(0, tcon, nls_codepage); - mutex_unlock(&tcon->ses->session_mutex); + mutex_unlock(&ses->session_mutex); cifs_dbg(FYI, "reconnect tcon rc = %d\n", rc); if (rc) { @@ -833,7 +850,9 @@ add_posix_context(struct kvec *iov, unsigned int *num_iovec, umode_t mode) */ int -SMB2_negotiate(const unsigned int xid, struct cifs_ses *ses) +SMB2_negotiate(const unsigned int xid, + struct cifs_ses *ses, + struct TCP_Server_Info *server) { struct smb_rqst rqst; struct smb2_negotiate_req *req; @@ -842,7 +861,6 @@ SMB2_negotiate(const unsigned int xid, struct cifs_ses *ses) struct kvec rsp_iov; int rc = 0; int resp_buftype; - struct TCP_Server_Info *server = cifs_ses_server(ses); int blob_offset, blob_length; char *security_blob; int flags = CIFS_NEG_OP; @@ -1221,6 +1239,7 @@ smb2_select_sectype(struct TCP_Server_Info *server, enum securityEnum requested) struct SMB2_sess_data { unsigned int xid; struct cifs_ses *ses; + struct TCP_Server_Info *server; struct nls_table *nls_cp; void (*func)(struct SMB2_sess_data *); int result; @@ -1242,9 +1261,10 @@ SMB2_sess_alloc_buffer(struct SMB2_sess_data *sess_data) { int rc; struct cifs_ses *ses = sess_data->ses; + struct TCP_Server_Info *server = sess_data->server; struct smb2_sess_setup_req *req; - struct TCP_Server_Info *server = cifs_ses_server(ses); unsigned int total_len; + bool is_binding = false; rc = smb2_plain_req_init(SMB2_SESSION_SETUP, NULL, server, (void **) &req, @@ -1252,11 +1272,16 @@ SMB2_sess_alloc_buffer(struct SMB2_sess_data *sess_data) if (rc) return rc; - if (sess_data->ses->binding) { - req->hdr.SessionId = cpu_to_le64(sess_data->ses->Suid); + spin_lock(&ses->chan_lock); + is_binding = !CIFS_ALL_CHANS_NEED_RECONNECT(ses); + spin_unlock(&ses->chan_lock); + + if (is_binding) { + req->hdr.SessionId = cpu_to_le64(ses->Suid); req->hdr.Flags |= SMB2_FLAGS_SIGNED; req->PreviousSessionId = 0; req->Flags = SMB2_SESSION_REQ_FLAG_BINDING; + cifs_dbg(FYI, "Binding to sess id: %llx\n", ses->Suid); } else { /* First session, not a reauthenticate */ req->hdr.SessionId = 0; @@ -1266,6 +1291,8 @@ SMB2_sess_alloc_buffer(struct SMB2_sess_data *sess_data) */ req->PreviousSessionId = cpu_to_le64(sess_data->previous_session); req->Flags = 0; /* MBZ */ + cifs_dbg(FYI, "Fresh session. Previous: %llx\n", + sess_data->previous_session); } /* enough to enable echos and oplocks and one max size write */ @@ -1325,7 +1352,7 @@ SMB2_sess_sendreceive(struct SMB2_sess_data *sess_data) /* BB add code to build os and lm fields */ rc = cifs_send_recv(sess_data->xid, sess_data->ses, - cifs_ses_server(sess_data->ses), + sess_data->server, &rqst, &sess_data->buf0_type, CIFS_LOG_ERROR | CIFS_SESS_OP, &rsp_iov); @@ -1340,11 +1367,11 @@ SMB2_sess_establish_session(struct SMB2_sess_data *sess_data) { int rc = 0; struct cifs_ses *ses = sess_data->ses; - struct TCP_Server_Info *server = cifs_ses_server(ses); + struct TCP_Server_Info *server = sess_data->server; mutex_lock(&server->srv_mutex); if (server->ops->generate_signingkey) { - rc = server->ops->generate_signingkey(ses); + rc = server->ops->generate_signingkey(ses, server); if (rc) { cifs_dbg(FYI, "SMB3 session key generation failed\n"); @@ -1359,14 +1386,6 @@ SMB2_sess_establish_session(struct SMB2_sess_data *sess_data) mutex_unlock(&server->srv_mutex); cifs_dbg(FYI, "SMB2/3 session established successfully\n"); - /* keep existing ses state if binding */ - if (!ses->binding) { - spin_lock(&GlobalMid_Lock); - ses->status = CifsGood; - ses->need_reconnect = false; - spin_unlock(&GlobalMid_Lock); - } - return rc; } @@ -1376,15 +1395,17 @@ SMB2_auth_kerberos(struct SMB2_sess_data *sess_data) { int rc; struct cifs_ses *ses = sess_data->ses; + struct TCP_Server_Info *server = sess_data->server; struct cifs_spnego_msg *msg; struct key *spnego_key = NULL; struct smb2_sess_setup_rsp *rsp = NULL; + bool is_binding = false; rc = SMB2_sess_alloc_buffer(sess_data); if (rc) goto out; - spnego_key = cifs_get_spnego_key(ses); + spnego_key = cifs_get_spnego_key(ses, server); if (IS_ERR(spnego_key)) { rc = PTR_ERR(spnego_key); if (rc == -ENOKEY) @@ -1405,8 +1426,12 @@ SMB2_auth_kerberos(struct SMB2_sess_data *sess_data) goto out_put_spnego_key; } + spin_lock(&ses->chan_lock); + is_binding = !CIFS_ALL_CHANS_NEED_RECONNECT(ses); + spin_unlock(&ses->chan_lock); + /* keep session key if binding */ - if (!ses->binding) { + if (!is_binding) { ses->auth_key.response = kmemdup(msg->data, msg->sesskey_len, GFP_KERNEL); if (!ses->auth_key.response) { @@ -1427,7 +1452,7 @@ SMB2_auth_kerberos(struct SMB2_sess_data *sess_data) rsp = (struct smb2_sess_setup_rsp *)sess_data->iov[0].iov_base; /* keep session id and flags if binding */ - if (!ses->binding) { + if (!is_binding) { ses->Suid = le64_to_cpu(rsp->hdr.SessionId); ses->session_flags = le16_to_cpu(rsp->SessionFlags); } @@ -1459,10 +1484,12 @@ SMB2_sess_auth_rawntlmssp_negotiate(struct SMB2_sess_data *sess_data) { int rc; struct cifs_ses *ses = sess_data->ses; + struct TCP_Server_Info *server = sess_data->server; struct smb2_sess_setup_rsp *rsp = NULL; unsigned char *ntlmssp_blob = NULL; bool use_spnego = false; /* else use raw ntlmssp */ u16 blob_length = 0; + bool is_binding = false; /* * If memory allocation is successful, caller of this function @@ -1479,8 +1506,8 @@ SMB2_sess_auth_rawntlmssp_negotiate(struct SMB2_sess_data *sess_data) if (rc) goto out_err; - rc = build_ntlmssp_negotiate_blob(&ntlmssp_blob, - &blob_length, ses, + rc = build_ntlmssp_smb3_negotiate_blob(&ntlmssp_blob, + &blob_length, ses, server, sess_data->nls_cp); if (rc) goto out_err; @@ -1519,8 +1546,12 @@ SMB2_sess_auth_rawntlmssp_negotiate(struct SMB2_sess_data *sess_data) cifs_dbg(FYI, "rawntlmssp session setup challenge phase\n"); + spin_lock(&ses->chan_lock); + is_binding = !CIFS_ALL_CHANS_NEED_RECONNECT(ses); + spin_unlock(&ses->chan_lock); + /* keep existing ses id and flags if binding */ - if (!ses->binding) { + if (!is_binding) { ses->Suid = le64_to_cpu(rsp->hdr.SessionId); ses->session_flags = le16_to_cpu(rsp->SessionFlags); } @@ -1545,11 +1576,13 @@ SMB2_sess_auth_rawntlmssp_authenticate(struct SMB2_sess_data *sess_data) { int rc; struct cifs_ses *ses = sess_data->ses; + struct TCP_Server_Info *server = sess_data->server; struct smb2_sess_setup_req *req; struct smb2_sess_setup_rsp *rsp = NULL; unsigned char *ntlmssp_blob = NULL; bool use_spnego = false; /* else use raw ntlmssp */ u16 blob_length = 0; + bool is_binding = false; rc = SMB2_sess_alloc_buffer(sess_data); if (rc) @@ -1558,8 +1591,9 @@ SMB2_sess_auth_rawntlmssp_authenticate(struct SMB2_sess_data *sess_data) req = (struct smb2_sess_setup_req *) sess_data->iov[0].iov_base; req->hdr.SessionId = cpu_to_le64(ses->Suid); - rc = build_ntlmssp_auth_blob(&ntlmssp_blob, &blob_length, ses, - sess_data->nls_cp); + rc = build_ntlmssp_auth_blob(&ntlmssp_blob, &blob_length, + ses, server, + sess_data->nls_cp); if (rc) { cifs_dbg(FYI, "build_ntlmssp_auth_blob failed %d\n", rc); goto out; @@ -1580,8 +1614,12 @@ SMB2_sess_auth_rawntlmssp_authenticate(struct SMB2_sess_data *sess_data) rsp = (struct smb2_sess_setup_rsp *)sess_data->iov[0].iov_base; + spin_lock(&ses->chan_lock); + is_binding = !CIFS_ALL_CHANS_NEED_RECONNECT(ses); + spin_unlock(&ses->chan_lock); + /* keep existing ses id and flags if binding */ - if (!ses->binding) { + if (!is_binding) { ses->Suid = le64_to_cpu(rsp->hdr.SessionId); ses->session_flags = le16_to_cpu(rsp->SessionFlags); } @@ -1612,11 +1650,13 @@ out: } static int -SMB2_select_sec(struct cifs_ses *ses, struct SMB2_sess_data *sess_data) +SMB2_select_sec(struct SMB2_sess_data *sess_data) { int type; + struct cifs_ses *ses = sess_data->ses; + struct TCP_Server_Info *server = sess_data->server; - type = smb2_select_sectype(cifs_ses_server(ses), ses->sectype); + type = smb2_select_sectype(server, ses->sectype); cifs_dbg(FYI, "sess setup type %d\n", type); if (type == Unspecified) { cifs_dbg(VFS, "Unable to select appropriate authentication method!\n"); @@ -1640,10 +1680,10 @@ SMB2_select_sec(struct cifs_ses *ses, struct SMB2_sess_data *sess_data) int SMB2_sess_setup(const unsigned int xid, struct cifs_ses *ses, + struct TCP_Server_Info *server, const struct nls_table *nls_cp) { int rc = 0; - struct TCP_Server_Info *server = cifs_ses_server(ses); struct SMB2_sess_data *sess_data; cifs_dbg(FYI, "Session Setup\n"); @@ -1657,15 +1697,17 @@ SMB2_sess_setup(const unsigned int xid, struct cifs_ses *ses, if (!sess_data) return -ENOMEM; - rc = SMB2_select_sec(ses, sess_data); - if (rc) - goto out; sess_data->xid = xid; sess_data->ses = ses; + sess_data->server = server; sess_data->buf0_type = CIFS_NO_BUFFER; sess_data->nls_cp = (struct nls_table *) nls_cp; sess_data->previous_session = ses->Suid; + rc = SMB2_select_sec(sess_data); + if (rc) + goto out; + /* * Initialize the session hash with the server one. */ @@ -1704,8 +1746,12 @@ SMB2_logoff(const unsigned int xid, struct cifs_ses *ses) return -EIO; /* no need to send SMB logoff if uid already closed due to reconnect */ - if (ses->need_reconnect) + spin_lock(&ses->chan_lock); + if (CIFS_ALL_CHANS_NEED_RECONNECT(ses)) { + spin_unlock(&ses->chan_lock); goto smb2_session_already_dead; + } + spin_unlock(&ses->chan_lock); rc = smb2_plain_req_init(SMB2_LOGOFF, NULL, ses->server, (void **) &req, &total_len); @@ -1867,8 +1913,6 @@ SMB2_tcon(const unsigned int xid, struct cifs_ses *ses, const char *tree, tcon->share_flags = le32_to_cpu(rsp->ShareFlags); tcon->capabilities = rsp->Capabilities; /* we keep caps little endian */ tcon->maximal_access = le32_to_cpu(rsp->MaximalAccess); - tcon->tidStatus = CifsGood; - tcon->need_reconnect = false; tcon->tid = le32_to_cpu(rsp->hdr.Id.SyncId.TreeId); strlcpy(tcon->treeName, tree, sizeof(tcon->treeName)); @@ -1913,8 +1957,13 @@ SMB2_tdis(const unsigned int xid, struct cifs_tcon *tcon) if (!ses || !(ses->server)) return -EIO; - if ((tcon->need_reconnect) || (tcon->ses->need_reconnect)) + spin_lock(&ses->chan_lock); + if ((tcon->need_reconnect) || + (CIFS_ALL_CHANS_NEED_RECONNECT(tcon->ses))) { + spin_unlock(&ses->chan_lock); return 0; + } + spin_unlock(&ses->chan_lock); close_cached_dir_lease(&tcon->crfid); @@ -2527,8 +2576,13 @@ alloc_path_with_tree_prefix(__le16 **out_path, int *out_size, int *out_len, cp = load_nls_default(); cifs_strtoUTF16(*out_path, treename, treename_len, cp); - UniStrcat(*out_path, sep); - UniStrcat(*out_path, path); + + /* Do not append the separator if the path is empty */ + if (path[0] != cpu_to_le16(0x0000)) { + UniStrcat(*out_path, sep); + UniStrcat(*out_path, path); + } + unload_nls(cp); return 0; @@ -3722,27 +3776,35 @@ void smb2_reconnect_server(struct work_struct *work) { struct TCP_Server_Info *server = container_of(work, struct TCP_Server_Info, reconnect.work); - struct cifs_ses *ses; + struct TCP_Server_Info *pserver; + struct cifs_ses *ses, *ses2; struct cifs_tcon *tcon, *tcon2; - struct list_head tmp_list; - int tcon_exist = false; + struct list_head tmp_list, tmp_ses_list; + bool tcon_exist = false, ses_exist = false; + bool tcon_selected = false; int rc; - int resched = false; + bool resched = false; + /* If server is a channel, select the primary channel */ + pserver = CIFS_SERVER_IS_CHAN(server) ? server->primary_server : server; /* Prevent simultaneous reconnects that can corrupt tcon->rlist list */ - mutex_lock(&server->reconnect_mutex); + mutex_lock(&pserver->reconnect_mutex); INIT_LIST_HEAD(&tmp_list); - cifs_dbg(FYI, "Need negotiate, reconnecting tcons\n"); + INIT_LIST_HEAD(&tmp_ses_list); + cifs_dbg(FYI, "Reconnecting tcons and channels\n"); spin_lock(&cifs_tcp_ses_lock); - list_for_each_entry(ses, &server->smb_ses_list, smb_ses_list) { + list_for_each_entry(ses, &pserver->smb_ses_list, smb_ses_list) { + + tcon_selected = false; + list_for_each_entry(tcon, &ses->tcon_list, tcon_list) { if (tcon->need_reconnect || tcon->need_reopen_files) { tcon->tc_count++; list_add_tail(&tcon->rlist, &tmp_list); - tcon_exist = true; + tcon_selected = tcon_exist = true; } } /* @@ -3751,15 +3813,27 @@ void smb2_reconnect_server(struct work_struct *work) */ if (ses->tcon_ipc && ses->tcon_ipc->need_reconnect) { list_add_tail(&ses->tcon_ipc->rlist, &tmp_list); - tcon_exist = true; + tcon_selected = tcon_exist = true; ses->ses_count++; } + /* + * handle the case where channel needs to reconnect + * binding session, but tcon is healthy (some other channel + * is active) + */ + spin_lock(&ses->chan_lock); + if (!tcon_selected && cifs_chan_needs_reconnect(ses, server)) { + list_add_tail(&ses->rlist, &tmp_ses_list); + ses_exist = true; + ses->ses_count++; + } + spin_unlock(&ses->chan_lock); } /* * Get the reference to server struct to be sure that the last call of * cifs_put_tcon() in the loop below won't release the server pointer. */ - if (tcon_exist) + if (tcon_exist || ses_exist) server->srv_count++; spin_unlock(&cifs_tcp_ses_lock); @@ -3777,13 +3851,41 @@ void smb2_reconnect_server(struct work_struct *work) cifs_put_tcon(tcon); } - cifs_dbg(FYI, "Reconnecting tcons finished\n"); + if (!ses_exist) + goto done; + + /* allocate a dummy tcon struct used for reconnect */ + tcon = kzalloc(sizeof(struct cifs_tcon), GFP_KERNEL); + if (!tcon) { + resched = true; + list_del_init(&ses->rlist); + cifs_put_smb_ses(ses); + goto done; + } + + tcon->tidStatus = CifsGood; + tcon->retry = false; + tcon->need_reconnect = false; + + /* now reconnect sessions for necessary channels */ + list_for_each_entry_safe(ses, ses2, &tmp_ses_list, rlist) { + tcon->ses = ses; + rc = smb2_reconnect(SMB2_INTERNAL_CMD, tcon, server); + if (rc) + resched = true; + list_del_init(&ses->rlist); + cifs_put_smb_ses(ses); + } + kfree(tcon); + +done: + cifs_dbg(FYI, "Reconnecting tcons and channels finished\n"); if (resched) queue_delayed_work(cifsiod_wq, &server->reconnect, 2 * HZ); - mutex_unlock(&server->reconnect_mutex); + mutex_unlock(&pserver->reconnect_mutex); /* now we can safely release srv struct */ - if (tcon_exist) + if (tcon_exist || ses_exist) cifs_put_tcp_session(server, 1); } @@ -3797,13 +3899,16 @@ SMB2_echo(struct TCP_Server_Info *server) .rq_nvec = 1 }; unsigned int total_len; - cifs_dbg(FYI, "In echo request\n"); + cifs_dbg(FYI, "In echo request for conn_id %lld\n", server->conn_id); + spin_lock(&cifs_tcp_ses_lock); if (server->tcpStatus == CifsNeedNegotiate) { + spin_unlock(&cifs_tcp_ses_lock); /* No need to send echo on newly established connections */ mod_delayed_work(cifsiod_wq, &server->reconnect, 0); return rc; } + spin_unlock(&cifs_tcp_ses_lock); rc = smb2_plain_req_init(SMB2_ECHO, NULL, server, (void **)&req, &total_len); diff --git a/fs/cifs/smb2proto.h b/fs/cifs/smb2proto.h index 096fada16ebd..4a7062fd1c26 100644 --- a/fs/cifs/smb2proto.h +++ b/fs/cifs/smb2proto.h @@ -123,8 +123,11 @@ extern void smb2_set_related(struct smb_rqst *rqst); * SMB2 Worker functions - most of protocol specific implementation details * are contained within these calls. */ -extern int SMB2_negotiate(const unsigned int xid, struct cifs_ses *ses); +extern int SMB2_negotiate(const unsigned int xid, + struct cifs_ses *ses, + struct TCP_Server_Info *server); extern int SMB2_sess_setup(const unsigned int xid, struct cifs_ses *ses, + struct TCP_Server_Info *server, const struct nls_table *nls_cp); extern int SMB2_logoff(const unsigned int xid, struct cifs_ses *ses); extern int SMB2_tcon(const unsigned int xid, struct cifs_ses *ses, @@ -276,6 +279,7 @@ extern void smb2_copy_fs_info_to_kstatfs( struct kstatfs *kst); extern int smb311_crypto_shash_allocate(struct TCP_Server_Info *server); extern int smb311_update_preauth_hash(struct cifs_ses *ses, + struct TCP_Server_Info *server, struct kvec *iov, int nvec); extern int smb2_query_info_compound(const unsigned int xid, struct cifs_tcon *tcon, diff --git a/fs/cifs/smb2transport.c b/fs/cifs/smb2transport.c index 2bf047b390a9..2af79093b78b 100644 --- a/fs/cifs/smb2transport.c +++ b/fs/cifs/smb2transport.c @@ -100,13 +100,16 @@ int smb2_get_sign_key(__u64 ses_id, struct TCP_Server_Info *server, u8 *key) goto out; found: - if (ses->binding) { + spin_lock(&ses->chan_lock); + if (cifs_chan_needs_reconnect(ses, server) && + !CIFS_ALL_CHANS_NEED_RECONNECT(ses)) { /* * If we are in the process of binding a new channel * to an existing session, use the master connection * session key */ memcpy(key, ses->smb3signingkey, SMB3_SIGN_KEY_SIZE); + spin_unlock(&ses->chan_lock); goto out; } @@ -118,9 +121,11 @@ found: chan = ses->chans + i; if (chan->server == server) { memcpy(key, chan->signkey, SMB3_SIGN_KEY_SIZE); + spin_unlock(&ses->chan_lock); goto out; } } + spin_unlock(&ses->chan_lock); cifs_dbg(VFS, "%s: Could not find channel signing key for session 0x%llx\n", @@ -390,12 +395,18 @@ struct derivation_triplet { static int generate_smb3signingkey(struct cifs_ses *ses, + struct TCP_Server_Info *server, const struct derivation_triplet *ptriplet) { int rc; -#ifdef CONFIG_CIFS_DEBUG_DUMP_KEYS - struct TCP_Server_Info *server = ses->server; -#endif + bool is_binding = false; + int chan_index = 0; + + spin_lock(&ses->chan_lock); + is_binding = !CIFS_ALL_CHANS_NEED_RECONNECT(ses); + chan_index = cifs_ses_get_chan_index(ses, server); + /* TODO: introduce ref counting for channels when the can be freed */ + spin_unlock(&ses->chan_lock); /* * All channels use the same encryption/decryption keys but @@ -407,10 +418,10 @@ generate_smb3signingkey(struct cifs_ses *ses, * master connection signing key stored in the session */ - if (ses->binding) { + if (is_binding) { rc = generate_key(ses, ptriplet->signing.label, ptriplet->signing.context, - cifs_ses_binding_channel(ses)->signkey, + ses->chans[chan_index].signkey, SMB3_SIGN_KEY_SIZE); if (rc) return rc; @@ -422,8 +433,11 @@ generate_smb3signingkey(struct cifs_ses *ses, if (rc) return rc; + /* safe to access primary channel, since it will never go away */ + spin_lock(&ses->chan_lock); memcpy(ses->chans[0].signkey, ses->smb3signingkey, SMB3_SIGN_KEY_SIZE); + spin_unlock(&ses->chan_lock); rc = generate_key(ses, ptriplet->encryption.label, ptriplet->encryption.context, @@ -470,7 +484,8 @@ generate_smb3signingkey(struct cifs_ses *ses, } int -generate_smb30signingkey(struct cifs_ses *ses) +generate_smb30signingkey(struct cifs_ses *ses, + struct TCP_Server_Info *server) { struct derivation_triplet triplet; @@ -494,11 +509,12 @@ generate_smb30signingkey(struct cifs_ses *ses) d->context.iov_base = "ServerOut"; d->context.iov_len = 10; - return generate_smb3signingkey(ses, &triplet); + return generate_smb3signingkey(ses, server, &triplet); } int -generate_smb311signingkey(struct cifs_ses *ses) +generate_smb311signingkey(struct cifs_ses *ses, + struct TCP_Server_Info *server) { struct derivation_triplet triplet; @@ -522,7 +538,7 @@ generate_smb311signingkey(struct cifs_ses *ses) d->context.iov_base = ses->preauth_sha_hash; d->context.iov_len = 64; - return generate_smb3signingkey(ses, &triplet); + return generate_smb3signingkey(ses, server, &triplet); } int @@ -624,8 +640,12 @@ smb2_sign_rqst(struct smb_rqst *rqst, struct TCP_Server_Info *server) if (!is_signed) return 0; - if (server->tcpStatus == CifsNeedNegotiate) + spin_lock(&cifs_tcp_ses_lock); + if (server->tcpStatus == CifsNeedNegotiate) { + spin_unlock(&cifs_tcp_ses_lock); return 0; + } + spin_unlock(&cifs_tcp_ses_lock); if (!is_binding && !server->session_estab) { strncpy(shdr->Signature, "BSRSPYL", 8); return 0; @@ -741,30 +761,41 @@ static int smb2_get_mid_entry(struct cifs_ses *ses, struct TCP_Server_Info *server, struct smb2_hdr *shdr, struct mid_q_entry **mid) { - if (server->tcpStatus == CifsExiting) + spin_lock(&cifs_tcp_ses_lock); + if (server->tcpStatus == CifsExiting) { + spin_unlock(&cifs_tcp_ses_lock); return -ENOENT; + } if (server->tcpStatus == CifsNeedReconnect) { + spin_unlock(&cifs_tcp_ses_lock); cifs_dbg(FYI, "tcp session dead - return to caller to retry\n"); return -EAGAIN; } if (server->tcpStatus == CifsNeedNegotiate && - shdr->Command != SMB2_NEGOTIATE) + shdr->Command != SMB2_NEGOTIATE) { + spin_unlock(&cifs_tcp_ses_lock); return -EAGAIN; + } if (ses->status == CifsNew) { if ((shdr->Command != SMB2_SESSION_SETUP) && - (shdr->Command != SMB2_NEGOTIATE)) + (shdr->Command != SMB2_NEGOTIATE)) { + spin_unlock(&cifs_tcp_ses_lock); return -EAGAIN; + } /* else ok - we are setting up session */ } if (ses->status == CifsExiting) { - if (shdr->Command != SMB2_LOGOFF) + if (shdr->Command != SMB2_LOGOFF) { + spin_unlock(&cifs_tcp_ses_lock); return -EAGAIN; + } /* else ok - we are shutting down the session */ } + spin_unlock(&cifs_tcp_ses_lock); *mid = smb2_mid_entry_alloc(shdr, server); if (*mid == NULL) @@ -837,9 +868,13 @@ smb2_setup_async_request(struct TCP_Server_Info *server, struct smb_rqst *rqst) (struct smb2_hdr *)rqst->rq_iov[0].iov_base; struct mid_q_entry *mid; + spin_lock(&cifs_tcp_ses_lock); if (server->tcpStatus == CifsNeedNegotiate && - shdr->Command != SMB2_NEGOTIATE) + shdr->Command != SMB2_NEGOTIATE) { + spin_unlock(&cifs_tcp_ses_lock); return ERR_PTR(-EAGAIN); + } + spin_unlock(&cifs_tcp_ses_lock); smb2_seq_num_into_buf(server, shdr); diff --git a/fs/cifs/transport.c b/fs/cifs/transport.c index 61ea3d3f95b4..8540f7c13eae 100644 --- a/fs/cifs/transport.c +++ b/fs/cifs/transport.c @@ -430,9 +430,10 @@ unmask: * be taken as the remainder of this one. We need to kill the * socket so the server throws away the partial SMB */ - spin_lock(&GlobalMid_Lock); - server->tcpStatus = CifsNeedReconnect; - spin_unlock(&GlobalMid_Lock); + spin_lock(&cifs_tcp_ses_lock); + if (server->tcpStatus != CifsExiting) + server->tcpStatus = CifsNeedReconnect; + spin_unlock(&cifs_tcp_ses_lock); trace_smb3_partial_send_reconnect(server->CurrentMid, server->conn_id, server->hostname); } @@ -578,10 +579,14 @@ wait_for_free_credits(struct TCP_Server_Info *server, const int num_credits, return -ERESTARTSYS; spin_lock(&server->req_lock); } else { + spin_unlock(&server->req_lock); + + spin_lock(&cifs_tcp_ses_lock); if (server->tcpStatus == CifsExiting) { - spin_unlock(&server->req_lock); + spin_unlock(&cifs_tcp_ses_lock); return -ENOENT; } + spin_unlock(&cifs_tcp_ses_lock); /* * For normal commands, reserve the last MAX_COMPOUND @@ -596,6 +601,7 @@ wait_for_free_credits(struct TCP_Server_Info *server, const int num_credits, * for servers that are slow to hand out credits on * new sessions. */ + spin_lock(&server->req_lock); if (!optype && num_credits == 1 && server->in_flight > 2 * MAX_COMPOUND && *credits <= MAX_COMPOUND) { @@ -723,28 +729,25 @@ cifs_wait_mtu_credits(struct TCP_Server_Info *server, unsigned int size, static int allocate_mid(struct cifs_ses *ses, struct smb_hdr *in_buf, struct mid_q_entry **ppmidQ) { - if (ses->server->tcpStatus == CifsExiting) { - return -ENOENT; - } - - if (ses->server->tcpStatus == CifsNeedReconnect) { - cifs_dbg(FYI, "tcp session dead - return to caller to retry\n"); - return -EAGAIN; - } - + spin_lock(&cifs_tcp_ses_lock); if (ses->status == CifsNew) { if ((in_buf->Command != SMB_COM_SESSION_SETUP_ANDX) && - (in_buf->Command != SMB_COM_NEGOTIATE)) + (in_buf->Command != SMB_COM_NEGOTIATE)) { + spin_unlock(&cifs_tcp_ses_lock); return -EAGAIN; + } /* else ok - we are setting up session */ } if (ses->status == CifsExiting) { /* check if SMB session is bad because we are setting it up */ - if (in_buf->Command != SMB_COM_LOGOFF_ANDX) + if (in_buf->Command != SMB_COM_LOGOFF_ANDX) { + spin_unlock(&cifs_tcp_ses_lock); return -EAGAIN; + } /* else ok - we are shutting down session */ } + spin_unlock(&cifs_tcp_ses_lock); *ppmidQ = AllocMidQEntry(in_buf, ses->server); if (*ppmidQ == NULL) @@ -1044,19 +1047,14 @@ struct TCP_Server_Info *cifs_pick_channel(struct cifs_ses *ses) if (!ses) return NULL; + /* round robin */ + index = (uint)atomic_inc_return(&ses->chan_seq); + spin_lock(&ses->chan_lock); - if (!ses->binding) { - /* round robin */ - if (ses->chan_count > 1) { - index = (uint)atomic_inc_return(&ses->chan_seq); - index %= ses->chan_count; - } - spin_unlock(&ses->chan_lock); - return ses->chans[index].server; - } else { - spin_unlock(&ses->chan_lock); - return cifs_ses_server(ses); - } + index %= ses->chan_count; + spin_unlock(&ses->chan_lock); + + return ses->chans[index].server; } int @@ -1084,8 +1082,12 @@ compound_send_recv(const unsigned int xid, struct cifs_ses *ses, return -EIO; } - if (server->tcpStatus == CifsExiting) + spin_lock(&cifs_tcp_ses_lock); + if (server->tcpStatus == CifsExiting) { + spin_unlock(&cifs_tcp_ses_lock); return -ENOENT; + } + spin_unlock(&cifs_tcp_ses_lock); /* * Wait for all the requests to become available. @@ -1188,12 +1190,17 @@ compound_send_recv(const unsigned int xid, struct cifs_ses *ses, /* * Compounding is never used during session establish. */ + spin_lock(&cifs_tcp_ses_lock); if ((ses->status == CifsNew) || (optype & CIFS_NEG_OP) || (optype & CIFS_SESS_OP)) { + spin_unlock(&cifs_tcp_ses_lock); + mutex_lock(&server->srv_mutex); - smb311_update_preauth_hash(ses, rqst[0].rq_iov, - rqst[0].rq_nvec); + smb311_update_preauth_hash(ses, server, rqst[0].rq_iov, rqst[0].rq_nvec); mutex_unlock(&server->srv_mutex); + + spin_lock(&cifs_tcp_ses_lock); } + spin_unlock(&cifs_tcp_ses_lock); for (i = 0; i < num_rqst; i++) { rc = wait_for_response(server, midQ[i]); @@ -1256,15 +1263,19 @@ compound_send_recv(const unsigned int xid, struct cifs_ses *ses, /* * Compounding is never used during session establish. */ + spin_lock(&cifs_tcp_ses_lock); if ((ses->status == CifsNew) || (optype & CIFS_NEG_OP) || (optype & CIFS_SESS_OP)) { struct kvec iov = { .iov_base = resp_iov[0].iov_base, .iov_len = resp_iov[0].iov_len }; + spin_unlock(&cifs_tcp_ses_lock); mutex_lock(&server->srv_mutex); - smb311_update_preauth_hash(ses, &iov, 1); + smb311_update_preauth_hash(ses, server, &iov, 1); mutex_unlock(&server->srv_mutex); + spin_lock(&cifs_tcp_ses_lock); } + spin_unlock(&cifs_tcp_ses_lock); out: /* @@ -1353,8 +1364,12 @@ SendReceive(const unsigned int xid, struct cifs_ses *ses, return -EIO; } - if (server->tcpStatus == CifsExiting) + spin_lock(&cifs_tcp_ses_lock); + if (server->tcpStatus == CifsExiting) { + spin_unlock(&cifs_tcp_ses_lock); return -ENOENT; + } + spin_unlock(&cifs_tcp_ses_lock); /* Ensure that we do not send more than 50 overlapping requests to the same server. We may make this configurable later or @@ -1494,8 +1509,12 @@ SendReceiveBlockingLock(const unsigned int xid, struct cifs_tcon *tcon, return -EIO; } - if (server->tcpStatus == CifsExiting) + spin_lock(&cifs_tcp_ses_lock); + if (server->tcpStatus == CifsExiting) { + spin_unlock(&cifs_tcp_ses_lock); return -ENOENT; + } + spin_unlock(&cifs_tcp_ses_lock); /* Ensure that we do not send more than 50 overlapping requests to the same server. We may make this configurable later or @@ -1553,10 +1572,12 @@ SendReceiveBlockingLock(const unsigned int xid, struct cifs_tcon *tcon, (server->tcpStatus != CifsNew))); /* Were we interrupted by a signal ? */ + spin_lock(&cifs_tcp_ses_lock); if ((rc == -ERESTARTSYS) && (midQ->mid_state == MID_REQUEST_SUBMITTED) && ((server->tcpStatus == CifsGood) || (server->tcpStatus == CifsNew))) { + spin_unlock(&cifs_tcp_ses_lock); if (in_buf->Command == SMB_COM_TRANSACTION2) { /* POSIX lock. We send a NT_CANCEL SMB to cause the @@ -1595,7 +1616,9 @@ SendReceiveBlockingLock(const unsigned int xid, struct cifs_tcon *tcon, /* We got the response - restart system call. */ rstart = 1; + spin_lock(&cifs_tcp_ses_lock); } + spin_unlock(&cifs_tcp_ses_lock); rc = cifs_sync_mid_result(midQ, server); if (rc != 0) diff --git a/fs/coredump.c b/fs/coredump.c index a6b3c196cdef..1c060c0a2d72 100644 --- a/fs/coredump.c +++ b/fs/coredump.c @@ -41,6 +41,7 @@ #include <linux/fs.h> #include <linux/path.h> #include <linux/timekeeping.h> +#include <linux/sysctl.h> #include <linux/uaccess.h> #include <asm/mmu_context.h> @@ -52,9 +53,9 @@ #include <trace/events/sched.h> -int core_uses_pid; -unsigned int core_pipe_limit; -char core_pattern[CORENAME_MAX_SIZE] = "core"; +static int core_uses_pid; +static unsigned int core_pipe_limit; +static char core_pattern[CORENAME_MAX_SIZE] = "core"; static int core_name_size = CORENAME_MAX_SIZE; struct core_name { @@ -62,8 +63,6 @@ struct core_name { int used, size; }; -/* The maximal length of core_pattern is also specified in sysctl.c */ - static int expand_corename(struct core_name *cn, int size) { char *corename = krealloc(cn->corename, size, GFP_KERNEL); @@ -347,13 +346,13 @@ out: return ispipe; } -static int zap_process(struct task_struct *start, int exit_code, int flags) +static int zap_process(struct task_struct *start, int exit_code) { struct task_struct *t; int nr = 0; /* ignore all signals except SIGKILL, see prepare_signal() */ - start->signal->flags = SIGNAL_GROUP_COREDUMP | flags; + start->signal->flags = SIGNAL_GROUP_EXIT; start->signal->group_exit_code = exit_code; start->signal->group_stop_count = 0; @@ -372,13 +371,13 @@ static int zap_process(struct task_struct *start, int exit_code, int flags) static int zap_threads(struct task_struct *tsk, struct core_state *core_state, int exit_code) { + struct signal_struct *signal = tsk->signal; int nr = -EAGAIN; spin_lock_irq(&tsk->sighand->siglock); - if (!signal_group_exit(tsk->signal)) { - tsk->signal->core_state = core_state; - tsk->signal->group_exit_task = tsk; - nr = zap_process(tsk, exit_code, 0); + if (!(signal->flags & SIGNAL_GROUP_EXIT) && !signal->group_exec_task) { + signal->core_state = core_state; + nr = zap_process(tsk, exit_code); clear_tsk_thread_flag(tsk, TIF_SIGPENDING); tsk->flags |= PF_DUMPCORE; atomic_set(&core_state->nr_threads, nr); @@ -426,8 +425,6 @@ static void coredump_finish(bool core_dumped) spin_lock_irq(¤t->sighand->siglock); if (core_dumped && !__fatal_signal_pending(current)) current->signal->group_exit_code |= 0x80; - current->signal->group_exit_task = NULL; - current->signal->flags = SIGNAL_GROUP_EXIT; next = current->signal->core_state->dumper.next; current->signal->core_state = NULL; spin_unlock_irq(¤t->sighand->siglock); @@ -895,6 +892,63 @@ int dump_align(struct coredump_params *cprm, int align) } EXPORT_SYMBOL(dump_align); +#ifdef CONFIG_SYSCTL + +void validate_coredump_safety(void) +{ + if (suid_dumpable == SUID_DUMP_ROOT && + core_pattern[0] != '/' && core_pattern[0] != '|') { + pr_warn( +"Unsafe core_pattern used with fs.suid_dumpable=2.\n" +"Pipe handler or fully qualified core dump path required.\n" +"Set kernel.core_pattern before fs.suid_dumpable.\n" + ); + } +} + +static int proc_dostring_coredump(struct ctl_table *table, int write, + void *buffer, size_t *lenp, loff_t *ppos) +{ + int error = proc_dostring(table, write, buffer, lenp, ppos); + + if (!error) + validate_coredump_safety(); + return error; +} + +static struct ctl_table coredump_sysctls[] = { + { + .procname = "core_uses_pid", + .data = &core_uses_pid, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec, + }, + { + .procname = "core_pattern", + .data = core_pattern, + .maxlen = CORENAME_MAX_SIZE, + .mode = 0644, + .proc_handler = proc_dostring_coredump, + }, + { + .procname = "core_pipe_limit", + .data = &core_pipe_limit, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = proc_dointvec, + }, + { } +}; + +static int __init init_fs_coredump_sysctls(void) +{ + register_sysctl_init("kernel", coredump_sysctls); + return 0; +} +fs_initcall(init_fs_coredump_sysctls); +#endif /* CONFIG_SYSCTL */ + /* * The purpose of always_dump_vma() is to make sure that special kernel mappings * that are useful for post-mortem analysis are included in every core dump. diff --git a/fs/dcache.c b/fs/dcache.c index cf871a81f4fd..c84269c6e8bf 100644 --- a/fs/dcache.c +++ b/fs/dcache.c @@ -115,10 +115,13 @@ static inline struct hlist_bl_head *in_lookup_hash(const struct dentry *parent, return in_lookup_hashtable + hash_32(hash, IN_LOOKUP_SHIFT); } - -/* Statistics gathering. */ -struct dentry_stat_t dentry_stat = { - .age_limit = 45, +struct dentry_stat_t { + long nr_dentry; + long nr_unused; + long age_limit; /* age in seconds */ + long want_pages; /* pages requested by system */ + long nr_negative; /* # of unused negative dentries */ + long dummy; /* Reserved for future use */ }; static DEFINE_PER_CPU(long, nr_dentry); @@ -126,6 +129,10 @@ static DEFINE_PER_CPU(long, nr_dentry_unused); static DEFINE_PER_CPU(long, nr_dentry_negative); #if defined(CONFIG_SYSCTL) && defined(CONFIG_PROC_FS) +/* Statistics gathering. */ +static struct dentry_stat_t dentry_stat = { + .age_limit = 45, +}; /* * Here we resort to our own counters instead of using generic per-cpu counters @@ -167,14 +174,32 @@ static long get_nr_dentry_negative(void) return sum < 0 ? 0 : sum; } -int proc_nr_dentry(struct ctl_table *table, int write, void *buffer, - size_t *lenp, loff_t *ppos) +static int proc_nr_dentry(struct ctl_table *table, int write, void *buffer, + size_t *lenp, loff_t *ppos) { dentry_stat.nr_dentry = get_nr_dentry(); dentry_stat.nr_unused = get_nr_dentry_unused(); dentry_stat.nr_negative = get_nr_dentry_negative(); return proc_doulongvec_minmax(table, write, buffer, lenp, ppos); } + +static struct ctl_table fs_dcache_sysctls[] = { + { + .procname = "dentry-state", + .data = &dentry_stat, + .maxlen = 6*sizeof(long), + .mode = 0444, + .proc_handler = proc_nr_dentry, + }, + { } +}; + +static int __init init_fs_dcache_sysctls(void) +{ + register_sysctl_init("fs", fs_dcache_sysctls); + return 0; +} +fs_initcall(init_fs_dcache_sysctls); #endif /* diff --git a/fs/eventpoll.c b/fs/eventpoll.c index 06f4c5ae1451..e2daa940ebce 100644 --- a/fs/eventpoll.c +++ b/fs/eventpoll.c @@ -307,7 +307,7 @@ static void unlist_file(struct epitems_head *head) static long long_zero; static long long_max = LONG_MAX; -struct ctl_table epoll_table[] = { +static struct ctl_table epoll_table[] = { { .procname = "max_user_watches", .data = &max_user_watches, @@ -319,6 +319,13 @@ struct ctl_table epoll_table[] = { }, { } }; + +static void __init epoll_sysctls_init(void) +{ + register_sysctl("fs/epoll", epoll_table); +} +#else +#define epoll_sysctls_init() do { } while (0) #endif /* CONFIG_SYSCTL */ static const struct file_operations eventpoll_fops; @@ -2378,6 +2385,7 @@ static int __init eventpoll_init(void) /* Allocates slab cache used to allocate "struct eppoll_entry" */ pwq_cache = kmem_cache_create("eventpoll_pwq", sizeof(struct eppoll_entry), 0, SLAB_PANIC|SLAB_ACCOUNT, NULL); + epoll_sysctls_init(); ephead_cache = kmem_cache_create("ep_head", sizeof(struct epitems_head), 0, SLAB_PANIC|SLAB_ACCOUNT, NULL); diff --git a/fs/exec.c b/fs/exec.c index 537d92c41105..79f2c9483302 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -65,6 +65,7 @@ #include <linux/vmalloc.h> #include <linux/io_uring.h> #include <linux/syscall_user_dispatch.h> +#include <linux/coredump.h> #include <linux/uaccess.h> #include <asm/mmu_context.h> @@ -1045,7 +1046,7 @@ static int de_thread(struct task_struct *tsk) * Kill all other threads in the thread group. */ spin_lock_irq(lock); - if (signal_group_exit(sig)) { + if ((sig->flags & SIGNAL_GROUP_EXIT) || sig->group_exec_task) { /* * Another group action in progress, just * return so that the signal is processed. @@ -1054,7 +1055,7 @@ static int de_thread(struct task_struct *tsk) return -EAGAIN; } - sig->group_exit_task = tsk; + sig->group_exec_task = tsk; sig->notify_count = zap_other_threads(tsk); if (!thread_group_leader(tsk)) sig->notify_count--; @@ -1082,7 +1083,7 @@ static int de_thread(struct task_struct *tsk) write_lock_irq(&tasklist_lock); /* * Do this under tasklist_lock to ensure that - * exit_notify() can't miss ->group_exit_task + * exit_notify() can't miss ->group_exec_task */ sig->notify_count = -1; if (likely(leader->exit_state)) @@ -1149,7 +1150,7 @@ static int de_thread(struct task_struct *tsk) release_task(leader); } - sig->group_exit_task = NULL; + sig->group_exec_task = NULL; sig->notify_count = 0; no_thread_group: @@ -1162,7 +1163,7 @@ no_thread_group: killed: /* protects against exit_notify() and __exit_signal() */ read_lock(&tasklist_lock); - sig->group_exit_task = NULL; + sig->group_exec_task = NULL; sig->notify_count = 0; read_unlock(&tasklist_lock); return -EAGAIN; @@ -1207,7 +1208,8 @@ static int unshare_sighand(struct task_struct *me) char *__get_task_comm(char *buf, size_t buf_size, struct task_struct *tsk) { task_lock(tsk); - strncpy(buf, tsk->comm, buf_size); + /* Always NUL terminated and zero-padded */ + strscpy_pad(buf, tsk->comm, buf_size); task_unlock(tsk); return buf; } @@ -1222,7 +1224,7 @@ void __set_task_comm(struct task_struct *tsk, const char *buf, bool exec) { task_lock(tsk); trace_task_rename(tsk, buf); - strlcpy(tsk->comm, buf, sizeof(tsk->comm)); + strscpy_pad(tsk->comm, buf, sizeof(tsk->comm)); task_unlock(tsk); perf_event_comm(tsk, exec); } @@ -1307,6 +1309,8 @@ int begin_new_exec(struct linux_binprm * bprm) */ force_uaccess_begin(); + if (me->flags & PF_KTHREAD) + free_kthread_struct(me); me->flags &= ~(PF_RANDOMIZE | PF_FORKNOEXEC | PF_KTHREAD | PF_NOFREEZE | PF_NO_SETAFFINITY); flush_thread(); @@ -2096,3 +2100,37 @@ COMPAT_SYSCALL_DEFINE5(execveat, int, fd, argv, envp, flags); } #endif + +#ifdef CONFIG_SYSCTL + +static int proc_dointvec_minmax_coredump(struct ctl_table *table, int write, + void *buffer, size_t *lenp, loff_t *ppos) +{ + int error = proc_dointvec_minmax(table, write, buffer, lenp, ppos); + + if (!error) + validate_coredump_safety(); + return error; +} + +static struct ctl_table fs_exec_sysctls[] = { + { + .procname = "suid_dumpable", + .data = &suid_dumpable, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec_minmax_coredump, + .extra1 = SYSCTL_ZERO, + .extra2 = SYSCTL_TWO, + }, + { } +}; + +static int __init init_fs_exec_sysctls(void) +{ + register_sysctl_init("fs", fs_exec_sysctls); + return 0; +} + +fs_initcall(init_fs_exec_sysctls); +#endif /* CONFIG_SYSCTL */ diff --git a/fs/exfat/balloc.c b/fs/exfat/balloc.c index cc5cffc4a769..03f142307174 100644 --- a/fs/exfat/balloc.c +++ b/fs/exfat/balloc.c @@ -105,7 +105,7 @@ int exfat_load_bitmap(struct super_block *sb) struct exfat_dentry *ep; struct buffer_head *bh; - ep = exfat_get_dentry(sb, &clu, i, &bh, NULL); + ep = exfat_get_dentry(sb, &clu, i, &bh); if (!ep) return -EIO; diff --git a/fs/exfat/dir.c b/fs/exfat/dir.c index cb1c0d8c1714..a27b55ec060a 100644 --- a/fs/exfat/dir.c +++ b/fs/exfat/dir.c @@ -64,7 +64,6 @@ static int exfat_readdir(struct inode *inode, loff_t *cpos, struct exfat_dir_ent { int i, dentries_per_clu, dentries_per_clu_bits = 0, num_ext; unsigned int type, clu_offset, max_dentries; - sector_t sector; struct exfat_chain dir, clu; struct exfat_uni_name uni_name; struct exfat_dentry *ep; @@ -115,7 +114,7 @@ static int exfat_readdir(struct inode *inode, loff_t *cpos, struct exfat_dir_ent i = dentry & (dentries_per_clu - 1); for ( ; i < dentries_per_clu; i++, dentry++) { - ep = exfat_get_dentry(sb, &clu, i, &bh, §or); + ep = exfat_get_dentry(sb, &clu, i, &bh); if (!ep) return -EIO; @@ -156,7 +155,7 @@ static int exfat_readdir(struct inode *inode, loff_t *cpos, struct exfat_dir_ent dir_entry->namebuf.lfnbuf_len); brelse(bh); - ep = exfat_get_dentry(sb, &clu, i + 1, &bh, NULL); + ep = exfat_get_dentry(sb, &clu, i + 1, &bh); if (!ep) return -EIO; dir_entry->size = @@ -445,7 +444,6 @@ int exfat_init_dir_entry(struct inode *inode, struct exfat_chain *p_dir, struct super_block *sb = inode->i_sb; struct exfat_sb_info *sbi = EXFAT_SB(sb); struct timespec64 ts = current_time(inode); - sector_t sector; struct exfat_dentry *ep; struct buffer_head *bh; @@ -453,7 +451,7 @@ int exfat_init_dir_entry(struct inode *inode, struct exfat_chain *p_dir, * We cannot use exfat_get_dentry_set here because file ep is not * initialized yet. */ - ep = exfat_get_dentry(sb, p_dir, entry, &bh, §or); + ep = exfat_get_dentry(sb, p_dir, entry, &bh); if (!ep) return -EIO; @@ -477,7 +475,7 @@ int exfat_init_dir_entry(struct inode *inode, struct exfat_chain *p_dir, exfat_update_bh(bh, IS_DIRSYNC(inode)); brelse(bh); - ep = exfat_get_dentry(sb, p_dir, entry + 1, &bh, §or); + ep = exfat_get_dentry(sb, p_dir, entry + 1, &bh); if (!ep) return -EIO; @@ -496,12 +494,11 @@ int exfat_update_dir_chksum(struct inode *inode, struct exfat_chain *p_dir, struct super_block *sb = inode->i_sb; int ret = 0; int i, num_entries; - sector_t sector; u16 chksum; struct exfat_dentry *ep, *fep; struct buffer_head *fbh, *bh; - fep = exfat_get_dentry(sb, p_dir, entry, &fbh, §or); + fep = exfat_get_dentry(sb, p_dir, entry, &fbh); if (!fep) return -EIO; @@ -509,7 +506,7 @@ int exfat_update_dir_chksum(struct inode *inode, struct exfat_chain *p_dir, chksum = exfat_calc_chksum16(fep, DENTRY_SIZE, 0, CS_DIR_ENTRY); for (i = 1; i < num_entries; i++) { - ep = exfat_get_dentry(sb, p_dir, entry + i, &bh, NULL); + ep = exfat_get_dentry(sb, p_dir, entry + i, &bh); if (!ep) { ret = -EIO; goto release_fbh; @@ -531,13 +528,12 @@ int exfat_init_ext_entry(struct inode *inode, struct exfat_chain *p_dir, { struct super_block *sb = inode->i_sb; int i; - sector_t sector; unsigned short *uniname = p_uniname->name; struct exfat_dentry *ep; struct buffer_head *bh; int sync = IS_DIRSYNC(inode); - ep = exfat_get_dentry(sb, p_dir, entry, &bh, §or); + ep = exfat_get_dentry(sb, p_dir, entry, &bh); if (!ep) return -EIO; @@ -545,7 +541,7 @@ int exfat_init_ext_entry(struct inode *inode, struct exfat_chain *p_dir, exfat_update_bh(bh, sync); brelse(bh); - ep = exfat_get_dentry(sb, p_dir, entry + 1, &bh, §or); + ep = exfat_get_dentry(sb, p_dir, entry + 1, &bh); if (!ep) return -EIO; @@ -555,7 +551,7 @@ int exfat_init_ext_entry(struct inode *inode, struct exfat_chain *p_dir, brelse(bh); for (i = EXFAT_FIRST_CLUSTER; i < num_entries; i++) { - ep = exfat_get_dentry(sb, p_dir, entry + i, &bh, §or); + ep = exfat_get_dentry(sb, p_dir, entry + i, &bh); if (!ep) return -EIO; @@ -574,12 +570,11 @@ int exfat_remove_entries(struct inode *inode, struct exfat_chain *p_dir, { struct super_block *sb = inode->i_sb; int i; - sector_t sector; struct exfat_dentry *ep; struct buffer_head *bh; for (i = order; i < num_entries; i++) { - ep = exfat_get_dentry(sb, p_dir, entry + i, &bh, §or); + ep = exfat_get_dentry(sb, p_dir, entry + i, &bh); if (!ep) return -EIO; @@ -656,8 +651,8 @@ static int exfat_walk_fat_chain(struct super_block *sb, return 0; } -int exfat_find_location(struct super_block *sb, struct exfat_chain *p_dir, - int entry, sector_t *sector, int *offset) +static int exfat_find_location(struct super_block *sb, struct exfat_chain *p_dir, + int entry, sector_t *sector, int *offset) { int ret; unsigned int off, clu = 0; @@ -717,8 +712,7 @@ static int exfat_dir_readahead(struct super_block *sb, sector_t sec) } struct exfat_dentry *exfat_get_dentry(struct super_block *sb, - struct exfat_chain *p_dir, int entry, struct buffer_head **bh, - sector_t *sector) + struct exfat_chain *p_dir, int entry, struct buffer_head **bh) { unsigned int dentries_per_page = EXFAT_B_TO_DEN(PAGE_SIZE); int off; @@ -740,8 +734,6 @@ struct exfat_dentry *exfat_get_dentry(struct super_block *sb, if (!*bh) return NULL; - if (sector) - *sector = sec; return (struct exfat_dentry *)((*bh)->b_data + off); } @@ -892,7 +884,7 @@ struct exfat_entry_set_cache *exfat_get_dentry_set(struct super_block *sb, es->bh[es->num_bh++] = bh; } - /* validiate cached dentries */ + /* validate cached dentries */ for (i = 1; i < num_entries; i++) { ep = exfat_get_dentry_cached(es, i); if (!exfat_validate_entry(exfat_get_entry_type(ep), &mode)) @@ -960,7 +952,7 @@ rewind: if (rewind && dentry == end_eidx) goto not_found; - ep = exfat_get_dentry(sb, &clu, i, &bh, NULL); + ep = exfat_get_dentry(sb, &clu, i, &bh); if (!ep) return -EIO; @@ -1145,7 +1137,7 @@ int exfat_count_ext_entries(struct super_block *sb, struct exfat_chain *p_dir, struct buffer_head *bh; for (i = 0, entry++; i < ep->dentry.file.num_ext; i++, entry++) { - ext_ep = exfat_get_dentry(sb, p_dir, entry, &bh, NULL); + ext_ep = exfat_get_dentry(sb, p_dir, entry, &bh); if (!ext_ep) return -EIO; @@ -1175,7 +1167,7 @@ int exfat_count_dir_entries(struct super_block *sb, struct exfat_chain *p_dir) while (clu.dir != EXFAT_EOF_CLUSTER) { for (i = 0; i < dentries_per_clu; i++) { - ep = exfat_get_dentry(sb, &clu, i, &bh, NULL); + ep = exfat_get_dentry(sb, &clu, i, &bh); if (!ep) return -EIO; entry_type = exfat_get_entry_type(ep); diff --git a/fs/exfat/exfat_fs.h b/fs/exfat/exfat_fs.h index 1d6da61157c9..619e5b4bed10 100644 --- a/fs/exfat/exfat_fs.h +++ b/fs/exfat/exfat_fs.h @@ -10,7 +10,6 @@ #include <linux/ratelimit.h> #include <linux/nls.h> -#define EXFAT_SUPER_MAGIC 0x2011BAB0UL #define EXFAT_ROOT_INO 1 #define EXFAT_CLUSTERS_UNTRACKED (~0u) @@ -459,11 +458,8 @@ int exfat_find_dir_entry(struct super_block *sb, struct exfat_inode_info *ei, struct exfat_chain *p_dir, struct exfat_uni_name *p_uniname, int num_entries, unsigned int type, struct exfat_hint *hint_opt); int exfat_alloc_new_dir(struct inode *inode, struct exfat_chain *clu); -int exfat_find_location(struct super_block *sb, struct exfat_chain *p_dir, - int entry, sector_t *sector, int *offset); struct exfat_dentry *exfat_get_dentry(struct super_block *sb, - struct exfat_chain *p_dir, int entry, struct buffer_head **bh, - sector_t *sector); + struct exfat_chain *p_dir, int entry, struct buffer_head **bh); struct exfat_dentry *exfat_get_dentry_cached(struct exfat_entry_set_cache *es, int num); struct exfat_entry_set_cache *exfat_get_dentry_set(struct super_block *sb, diff --git a/fs/exfat/fatent.c b/fs/exfat/fatent.c index e949e563443c..a3464e56a7e1 100644 --- a/fs/exfat/fatent.c +++ b/fs/exfat/fatent.c @@ -84,9 +84,7 @@ int exfat_ent_set(struct super_block *sb, unsigned int loc, static inline bool is_valid_cluster(struct exfat_sb_info *sbi, unsigned int clus) { - if (clus < EXFAT_FIRST_CLUSTER || sbi->num_clusters <= clus) - return false; - return true; + return clus >= EXFAT_FIRST_CLUSTER && clus < sbi->num_clusters; } int exfat_ent_get(struct super_block *sb, unsigned int loc, diff --git a/fs/exfat/file.c b/fs/exfat/file.c index 6af0191b648f..d890fd34bb2d 100644 --- a/fs/exfat/file.c +++ b/fs/exfat/file.c @@ -110,8 +110,7 @@ int __exfat_truncate(struct inode *inode, loff_t new_size) exfat_set_volume_dirty(sb); num_clusters_new = EXFAT_B_TO_CLU_ROUND_UP(i_size_read(inode), sbi); - num_clusters_phys = - EXFAT_B_TO_CLU_ROUND_UP(EXFAT_I(inode)->i_size_ondisk, sbi); + num_clusters_phys = EXFAT_B_TO_CLU_ROUND_UP(ei->i_size_ondisk, sbi); exfat_chain_set(&clu, ei->start_clu, num_clusters_phys, ei->flags); @@ -228,12 +227,13 @@ void exfat_truncate(struct inode *inode, loff_t size) { struct super_block *sb = inode->i_sb; struct exfat_sb_info *sbi = EXFAT_SB(sb); + struct exfat_inode_info *ei = EXFAT_I(inode); unsigned int blocksize = i_blocksize(inode); loff_t aligned_size; int err; mutex_lock(&sbi->s_lock); - if (EXFAT_I(inode)->start_clu == 0) { + if (ei->start_clu == 0) { /* * Empty start_clu != ~0 (not allocated) */ @@ -251,8 +251,8 @@ void exfat_truncate(struct inode *inode, loff_t size) else mark_inode_dirty(inode); - inode->i_blocks = ((i_size_read(inode) + (sbi->cluster_size - 1)) & - ~(sbi->cluster_size - 1)) >> inode->i_blkbits; + inode->i_blocks = round_up(i_size_read(inode), sbi->cluster_size) >> + inode->i_blkbits; write_size: aligned_size = i_size_read(inode); if (aligned_size & (blocksize - 1)) { @@ -260,11 +260,11 @@ write_size: aligned_size++; } - if (EXFAT_I(inode)->i_size_ondisk > i_size_read(inode)) - EXFAT_I(inode)->i_size_ondisk = aligned_size; + if (ei->i_size_ondisk > i_size_read(inode)) + ei->i_size_ondisk = aligned_size; - if (EXFAT_I(inode)->i_size_aligned > i_size_read(inode)) - EXFAT_I(inode)->i_size_aligned = aligned_size; + if (ei->i_size_aligned > i_size_read(inode)) + ei->i_size_aligned = aligned_size; mutex_unlock(&sbi->s_lock); } diff --git a/fs/exfat/inode.c b/fs/exfat/inode.c index 1c7aa1ea4724..df805bd05508 100644 --- a/fs/exfat/inode.c +++ b/fs/exfat/inode.c @@ -31,7 +31,7 @@ static int __exfat_write_inode(struct inode *inode, int sync) return 0; /* - * If the indode is already unlinked, there is no need for updating it. + * If the inode is already unlinked, there is no need for updating it. */ if (ei->dir.dir == DIR_DELETED) return 0; @@ -114,10 +114,9 @@ static int exfat_map_cluster(struct inode *inode, unsigned int clu_offset, unsigned int local_clu_offset = clu_offset; unsigned int num_to_be_allocated = 0, num_clusters = 0; - if (EXFAT_I(inode)->i_size_ondisk > 0) + if (ei->i_size_ondisk > 0) num_clusters = - EXFAT_B_TO_CLU_ROUND_UP(EXFAT_I(inode)->i_size_ondisk, - sbi); + EXFAT_B_TO_CLU_ROUND_UP(ei->i_size_ondisk, sbi); if (clu_offset >= num_clusters) num_to_be_allocated = clu_offset - num_clusters + 1; @@ -416,10 +415,10 @@ static int exfat_write_end(struct file *file, struct address_space *mapping, err = generic_write_end(file, mapping, pos, len, copied, pagep, fsdata); - if (EXFAT_I(inode)->i_size_aligned < i_size_read(inode)) { + if (ei->i_size_aligned < i_size_read(inode)) { exfat_fs_error(inode->i_sb, "invalid size(size(%llu) > aligned(%llu)\n", - i_size_read(inode), EXFAT_I(inode)->i_size_aligned); + i_size_read(inode), ei->i_size_aligned); return -EIO; } @@ -603,8 +602,8 @@ static int exfat_fill_inode(struct inode *inode, struct exfat_dir_entry *info) exfat_save_attr(inode, info->attr); - inode->i_blocks = ((i_size_read(inode) + (sbi->cluster_size - 1)) & - ~((loff_t)sbi->cluster_size - 1)) >> inode->i_blkbits; + inode->i_blocks = round_up(i_size_read(inode), sbi->cluster_size) >> + inode->i_blkbits; inode->i_mtime = info->mtime; inode->i_ctime = info->mtime; ei->i_crtime = info->crtime; diff --git a/fs/exfat/misc.c b/fs/exfat/misc.c index d34e6193258d..d5bd8e6d9741 100644 --- a/fs/exfat/misc.c +++ b/fs/exfat/misc.c @@ -10,6 +10,7 @@ #include <linux/fs.h> #include <linux/slab.h> #include <linux/buffer_head.h> +#include <linux/blk_types.h> #include "exfat_raw.h" #include "exfat_fs.h" @@ -180,7 +181,7 @@ int exfat_update_bhs(struct buffer_head **bhs, int nr_bhs, int sync) set_buffer_uptodate(bhs[i]); mark_buffer_dirty(bhs[i]); if (sync) - write_dirty_buffer(bhs[i], 0); + write_dirty_buffer(bhs[i], REQ_SYNC); } for (i = 0; i < nr_bhs && sync; i++) { diff --git a/fs/exfat/namei.c b/fs/exfat/namei.c index 24b41103d1cc..af4eb39cc0c3 100644 --- a/fs/exfat/namei.c +++ b/fs/exfat/namei.c @@ -229,7 +229,7 @@ static int exfat_search_empty_slot(struct super_block *sb, i = dentry & (dentries_per_clu - 1); for (; i < dentries_per_clu; i++, dentry++) { - ep = exfat_get_dentry(sb, &clu, i, &bh, NULL); + ep = exfat_get_dentry(sb, &clu, i, &bh); if (!ep) return -EIO; type = exfat_get_entry_type(ep); @@ -306,7 +306,6 @@ static int exfat_find_empty_entry(struct inode *inode, { int dentry; unsigned int ret, last_clu; - sector_t sector; loff_t size = 0; struct exfat_chain clu; struct exfat_dentry *ep = NULL; @@ -379,7 +378,7 @@ static int exfat_find_empty_entry(struct inode *inode, struct buffer_head *bh; ep = exfat_get_dentry(sb, - &(ei->dir), ei->entry + 1, &bh, §or); + &(ei->dir), ei->entry + 1, &bh); if (!ep) return -EIO; @@ -395,9 +394,9 @@ static int exfat_find_empty_entry(struct inode *inode, /* directory inode should be updated in here */ i_size_write(inode, size); - EXFAT_I(inode)->i_size_ondisk += sbi->cluster_size; - EXFAT_I(inode)->i_size_aligned += sbi->cluster_size; - EXFAT_I(inode)->flags = p_dir->flags; + ei->i_size_ondisk += sbi->cluster_size; + ei->i_size_aligned += sbi->cluster_size; + ei->flags = p_dir->flags; inode->i_blocks += 1 << sbi->sect_per_clus_bits; } @@ -779,7 +778,6 @@ static int exfat_unlink(struct inode *dir, struct dentry *dentry) struct inode *inode = dentry->d_inode; struct exfat_inode_info *ei = EXFAT_I(inode); struct buffer_head *bh; - sector_t sector; int num_entries, entry, err = 0; mutex_lock(&EXFAT_SB(sb)->s_lock); @@ -791,7 +789,7 @@ static int exfat_unlink(struct inode *dir, struct dentry *dentry) goto unlock; } - ep = exfat_get_dentry(sb, &cdir, entry, &bh, §or); + ep = exfat_get_dentry(sb, &cdir, entry, &bh); if (!ep) { err = -EIO; goto unlock; @@ -895,7 +893,7 @@ static int exfat_check_dir_empty(struct super_block *sb, while (clu.dir != EXFAT_EOF_CLUSTER) { for (i = 0; i < dentries_per_clu; i++) { - ep = exfat_get_dentry(sb, &clu, i, &bh, NULL); + ep = exfat_get_dentry(sb, &clu, i, &bh); if (!ep) return -EIO; type = exfat_get_entry_type(ep); @@ -932,7 +930,6 @@ static int exfat_rmdir(struct inode *dir, struct dentry *dentry) struct exfat_sb_info *sbi = EXFAT_SB(sb); struct exfat_inode_info *ei = EXFAT_I(inode); struct buffer_head *bh; - sector_t sector; int num_entries, entry, err; mutex_lock(&EXFAT_SB(inode->i_sb)->s_lock); @@ -957,7 +954,7 @@ static int exfat_rmdir(struct inode *dir, struct dentry *dentry) goto unlock; } - ep = exfat_get_dentry(sb, &cdir, entry, &bh, §or); + ep = exfat_get_dentry(sb, &cdir, entry, &bh); if (!ep) { err = -EIO; goto unlock; @@ -1005,13 +1002,12 @@ static int exfat_rename_file(struct inode *inode, struct exfat_chain *p_dir, struct exfat_inode_info *ei) { int ret, num_old_entries, num_new_entries; - sector_t sector_old, sector_new; struct exfat_dentry *epold, *epnew; struct super_block *sb = inode->i_sb; struct buffer_head *new_bh, *old_bh; int sync = IS_DIRSYNC(inode); - epold = exfat_get_dentry(sb, p_dir, oldentry, &old_bh, §or_old); + epold = exfat_get_dentry(sb, p_dir, oldentry, &old_bh); if (!epold) return -EIO; @@ -1032,8 +1028,7 @@ static int exfat_rename_file(struct inode *inode, struct exfat_chain *p_dir, if (newentry < 0) return newentry; /* -EIO or -ENOSPC */ - epnew = exfat_get_dentry(sb, p_dir, newentry, &new_bh, - §or_new); + epnew = exfat_get_dentry(sb, p_dir, newentry, &new_bh); if (!epnew) return -EIO; @@ -1046,12 +1041,10 @@ static int exfat_rename_file(struct inode *inode, struct exfat_chain *p_dir, brelse(old_bh); brelse(new_bh); - epold = exfat_get_dentry(sb, p_dir, oldentry + 1, &old_bh, - §or_old); + epold = exfat_get_dentry(sb, p_dir, oldentry + 1, &old_bh); if (!epold) return -EIO; - epnew = exfat_get_dentry(sb, p_dir, newentry + 1, &new_bh, - §or_new); + epnew = exfat_get_dentry(sb, p_dir, newentry + 1, &new_bh); if (!epnew) { brelse(old_bh); return -EIO; @@ -1093,12 +1086,11 @@ static int exfat_move_file(struct inode *inode, struct exfat_chain *p_olddir, struct exfat_uni_name *p_uniname, struct exfat_inode_info *ei) { int ret, newentry, num_new_entries, num_old_entries; - sector_t sector_mov, sector_new; struct exfat_dentry *epmov, *epnew; struct super_block *sb = inode->i_sb; struct buffer_head *mov_bh, *new_bh; - epmov = exfat_get_dentry(sb, p_olddir, oldentry, &mov_bh, §or_mov); + epmov = exfat_get_dentry(sb, p_olddir, oldentry, &mov_bh); if (!epmov) return -EIO; @@ -1116,7 +1108,7 @@ static int exfat_move_file(struct inode *inode, struct exfat_chain *p_olddir, if (newentry < 0) return newentry; /* -EIO or -ENOSPC */ - epnew = exfat_get_dentry(sb, p_newdir, newentry, &new_bh, §or_new); + epnew = exfat_get_dentry(sb, p_newdir, newentry, &new_bh); if (!epnew) return -EIO; @@ -1129,12 +1121,10 @@ static int exfat_move_file(struct inode *inode, struct exfat_chain *p_olddir, brelse(mov_bh); brelse(new_bh); - epmov = exfat_get_dentry(sb, p_olddir, oldentry + 1, &mov_bh, - §or_mov); + epmov = exfat_get_dentry(sb, p_olddir, oldentry + 1, &mov_bh); if (!epmov) return -EIO; - epnew = exfat_get_dentry(sb, p_newdir, newentry + 1, &new_bh, - §or_new); + epnew = exfat_get_dentry(sb, p_newdir, newentry + 1, &new_bh); if (!epnew) { brelse(mov_bh); return -EIO; @@ -1216,7 +1206,7 @@ static int __exfat_rename(struct inode *old_parent_inode, exfat_chain_dup(&olddir, &ei->dir); dentry = ei->entry; - ep = exfat_get_dentry(sb, &olddir, dentry, &old_bh, NULL); + ep = exfat_get_dentry(sb, &olddir, dentry, &old_bh); if (!ep) { ret = -EIO; goto out; @@ -1237,7 +1227,7 @@ static int __exfat_rename(struct inode *old_parent_inode, p_dir = &(new_ei->dir); new_entry = new_ei->entry; - ep = exfat_get_dentry(sb, p_dir, new_entry, &new_bh, NULL); + ep = exfat_get_dentry(sb, p_dir, new_entry, &new_bh); if (!ep) goto out; @@ -1277,7 +1267,7 @@ static int __exfat_rename(struct inode *old_parent_inode, if (!ret && new_inode) { /* delete entries of new_dir */ - ep = exfat_get_dentry(sb, p_dir, new_entry, &new_bh, NULL); + ep = exfat_get_dentry(sb, p_dir, new_entry, &new_bh); if (!ep) { ret = -EIO; goto del_out; diff --git a/fs/exfat/nls.c b/fs/exfat/nls.c index 314d5407a1be..ef115e673406 100644 --- a/fs/exfat/nls.c +++ b/fs/exfat/nls.c @@ -761,7 +761,7 @@ int exfat_create_upcase_table(struct super_block *sb) while (clu.dir != EXFAT_EOF_CLUSTER) { for (i = 0; i < sbi->dentries_per_clu; i++) { - ep = exfat_get_dentry(sb, &clu, i, &bh, NULL); + ep = exfat_get_dentry(sb, &clu, i, &bh); if (!ep) return -EIO; diff --git a/fs/exfat/super.c b/fs/exfat/super.c index 5539ffc20d16..8c9fb7dcec16 100644 --- a/fs/exfat/super.c +++ b/fs/exfat/super.c @@ -17,6 +17,7 @@ #include <linux/iversion.h> #include <linux/nls.h> #include <linux/buffer_head.h> +#include <linux/magic.h> #include "exfat_raw.h" #include "exfat_fs.h" @@ -364,11 +365,11 @@ static int exfat_read_root(struct inode *inode) inode->i_op = &exfat_dir_inode_operations; inode->i_fop = &exfat_dir_operations; - inode->i_blocks = ((i_size_read(inode) + (sbi->cluster_size - 1)) - & ~(sbi->cluster_size - 1)) >> inode->i_blkbits; - EXFAT_I(inode)->i_pos = ((loff_t)sbi->root_dir << 32) | 0xffffffff; - EXFAT_I(inode)->i_size_aligned = i_size_read(inode); - EXFAT_I(inode)->i_size_ondisk = i_size_read(inode); + inode->i_blocks = round_up(i_size_read(inode), sbi->cluster_size) >> + inode->i_blkbits; + ei->i_pos = ((loff_t)sbi->root_dir << 32) | 0xffffffff; + ei->i_size_aligned = i_size_read(inode); + ei->i_size_ondisk = i_size_read(inode); exfat_save_attr(inode, ATTR_SUBDIR); inode->i_mtime = inode->i_atime = inode->i_ctime = ei->i_crtime = diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index 1077ce7e189f..74c91da585d7 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -27,8 +27,8 @@ #include <linux/slab.h> #include <linux/uaccess.h> #include <linux/fiemap.h> -#include <linux/backing-dev.h> #include <linux/iomap.h> +#include <linux/sched/mm.h> #include "ext4_jbd2.h" #include "ext4_extents.h" #include "xattr.h" @@ -4404,8 +4404,7 @@ retry: err = ext4_es_remove_extent(inode, last_block, EXT_MAX_BLOCKS - last_block); if (err == -ENOMEM) { - cond_resched(); - congestion_wait(BLK_RW_ASYNC, HZ/50); + memalloc_retry_wait(GFP_ATOMIC); goto retry; } if (err) @@ -4413,8 +4412,7 @@ retry: retry_remove_space: err = ext4_ext_remove_space(inode, last_block, EXT_MAX_BLOCKS - 1); if (err == -ENOMEM) { - cond_resched(); - congestion_wait(BLK_RW_ASYNC, HZ/50); + memalloc_retry_wait(GFP_ATOMIC); goto retry_remove_space; } return err; diff --git a/fs/ext4/inline.c b/fs/ext4/inline.c index 39a1ab129fdc..635bcf68a67e 100644 --- a/fs/ext4/inline.c +++ b/fs/ext4/inline.c @@ -7,7 +7,7 @@ #include <linux/iomap.h> #include <linux/fiemap.h> #include <linux/iversion.h> -#include <linux/backing-dev.h> +#include <linux/sched/mm.h> #include "ext4_jbd2.h" #include "ext4.h" @@ -1929,8 +1929,7 @@ int ext4_inline_data_truncate(struct inode *inode, int *has_inline) retry: err = ext4_es_remove_extent(inode, 0, EXT_MAX_BLOCKS); if (err == -ENOMEM) { - cond_resched(); - congestion_wait(BLK_RW_ASYNC, HZ/50); + memalloc_retry_wait(GFP_ATOMIC); goto retry; } if (err) diff --git a/fs/ext4/mballoc.c b/fs/ext4/mballoc.c index cf2fd9fc7d98..9f86dd947032 100644 --- a/fs/ext4/mballoc.c +++ b/fs/ext4/mballoc.c @@ -2834,7 +2834,7 @@ out: static void *ext4_mb_seq_groups_start(struct seq_file *seq, loff_t *pos) { - struct super_block *sb = PDE_DATA(file_inode(seq->file)); + struct super_block *sb = pde_data(file_inode(seq->file)); ext4_group_t group; if (*pos < 0 || *pos >= ext4_get_groups_count(sb)) @@ -2845,7 +2845,7 @@ static void *ext4_mb_seq_groups_start(struct seq_file *seq, loff_t *pos) static void *ext4_mb_seq_groups_next(struct seq_file *seq, void *v, loff_t *pos) { - struct super_block *sb = PDE_DATA(file_inode(seq->file)); + struct super_block *sb = pde_data(file_inode(seq->file)); ext4_group_t group; ++*pos; @@ -2857,7 +2857,7 @@ static void *ext4_mb_seq_groups_next(struct seq_file *seq, void *v, loff_t *pos) static int ext4_mb_seq_groups_show(struct seq_file *seq, void *v) { - struct super_block *sb = PDE_DATA(file_inode(seq->file)); + struct super_block *sb = pde_data(file_inode(seq->file)); ext4_group_t group = (ext4_group_t) ((unsigned long) v); int i; int err, buddy_loaded = 0; @@ -2985,7 +2985,7 @@ int ext4_seq_mb_stats_show(struct seq_file *seq, void *offset) static void *ext4_mb_seq_structs_summary_start(struct seq_file *seq, loff_t *pos) __acquires(&EXT4_SB(sb)->s_mb_rb_lock) { - struct super_block *sb = PDE_DATA(file_inode(seq->file)); + struct super_block *sb = pde_data(file_inode(seq->file)); unsigned long position; read_lock(&EXT4_SB(sb)->s_mb_rb_lock); @@ -2998,7 +2998,7 @@ __acquires(&EXT4_SB(sb)->s_mb_rb_lock) static void *ext4_mb_seq_structs_summary_next(struct seq_file *seq, void *v, loff_t *pos) { - struct super_block *sb = PDE_DATA(file_inode(seq->file)); + struct super_block *sb = pde_data(file_inode(seq->file)); unsigned long position; ++*pos; @@ -3010,7 +3010,7 @@ static void *ext4_mb_seq_structs_summary_next(struct seq_file *seq, void *v, lof static int ext4_mb_seq_structs_summary_show(struct seq_file *seq, void *v) { - struct super_block *sb = PDE_DATA(file_inode(seq->file)); + struct super_block *sb = pde_data(file_inode(seq->file)); struct ext4_sb_info *sbi = EXT4_SB(sb); unsigned long position = ((unsigned long) v); struct ext4_group_info *grp; @@ -3058,7 +3058,7 @@ static int ext4_mb_seq_structs_summary_show(struct seq_file *seq, void *v) static void ext4_mb_seq_structs_summary_stop(struct seq_file *seq, void *v) __releases(&EXT4_SB(sb)->s_mb_rb_lock) { - struct super_block *sb = PDE_DATA(file_inode(seq->file)); + struct super_block *sb = pde_data(file_inode(seq->file)); read_unlock(&EXT4_SB(sb)->s_mb_rb_lock); } diff --git a/fs/ext4/page-io.c b/fs/ext4/page-io.c index 9cb261714991..1d370364230e 100644 --- a/fs/ext4/page-io.c +++ b/fs/ext4/page-io.c @@ -24,7 +24,7 @@ #include <linux/kernel.h> #include <linux/slab.h> #include <linux/mm.h> -#include <linux/backing-dev.h> +#include <linux/sched/mm.h> #include "ext4_jbd2.h" #include "xattr.h" @@ -523,12 +523,13 @@ int ext4_bio_write_page(struct ext4_io_submit *io, ret = PTR_ERR(bounce_page); if (ret == -ENOMEM && (io->io_bio || wbc->sync_mode == WB_SYNC_ALL)) { - gfp_flags = GFP_NOFS; + gfp_t new_gfp_flags = GFP_NOFS; if (io->io_bio) ext4_io_submit(io); else - gfp_flags |= __GFP_NOFAIL; - congestion_wait(BLK_RW_ASYNC, HZ/50); + new_gfp_flags |= __GFP_NOFAIL; + memalloc_retry_wait(gfp_flags); + gfp_flags = new_gfp_flags; goto retry_encrypt; } diff --git a/fs/ext4/readpage.c b/fs/ext4/readpage.c index 3db923403505..4cd62f1d848c 100644 --- a/fs/ext4/readpage.c +++ b/fs/ext4/readpage.c @@ -43,7 +43,6 @@ #include <linux/writeback.h> #include <linux/backing-dev.h> #include <linux/pagevec.h> -#include <linux/cleancache.h> #include "ext4.h" @@ -350,11 +349,6 @@ int ext4_mpage_readpages(struct inode *inode, } else if (fully_mapped) { SetPageMappedToDisk(page); } - if (fully_mapped && blocks_per_page == 1 && - !PageUptodate(page) && cleancache_get_page(page) == 0) { - SetPageUptodate(page); - goto confused; - } /* * This page will go to BIO. Do we need to send this diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 0343f682504d..eee0d9ebfa6c 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -39,7 +39,6 @@ #include <linux/log2.h> #include <linux/crc16.h> #include <linux/dax.h> -#include <linux/cleancache.h> #include <linux/uaccess.h> #include <linux/iversion.h> #include <linux/unicode.h> @@ -1966,29 +1965,22 @@ static const struct mount_opts { static const struct ext4_sb_encodings { __u16 magic; char *name; - char *version; + unsigned int version; } ext4_sb_encoding_map[] = { - {EXT4_ENC_UTF8_12_1, "utf8", "12.1.0"}, + {EXT4_ENC_UTF8_12_1, "utf8", UNICODE_AGE(12, 1, 0)}, }; -static int ext4_sb_read_encoding(const struct ext4_super_block *es, - const struct ext4_sb_encodings **encoding, - __u16 *flags) +static const struct ext4_sb_encodings * +ext4_sb_read_encoding(const struct ext4_super_block *es) { __u16 magic = le16_to_cpu(es->s_encoding); int i; for (i = 0; i < ARRAY_SIZE(ext4_sb_encoding_map); i++) if (magic == ext4_sb_encoding_map[i].magic) - break; - - if (i >= ARRAY_SIZE(ext4_sb_encoding_map)) - return -EINVAL; - - *encoding = &ext4_sb_encoding_map[i]; - *flags = le16_to_cpu(es->s_encoding_flags); + return &ext4_sb_encoding_map[i]; - return 0; + return NULL; } #endif @@ -3156,8 +3148,6 @@ done: EXT4_BLOCKS_PER_GROUP(sb), EXT4_INODES_PER_GROUP(sb), sbi->s_mount_opt, sbi->s_mount_opt2); - - cleancache_init_fs(sb); return err; } @@ -4624,10 +4614,10 @@ static int __ext4_fill_super(struct fs_context *fc, struct super_block *sb) if (ext4_has_feature_casefold(sb) && !sb->s_encoding) { const struct ext4_sb_encodings *encoding_info; struct unicode_map *encoding; - __u16 encoding_flags; + __u16 encoding_flags = le16_to_cpu(es->s_encoding_flags); - if (ext4_sb_read_encoding(es, &encoding_info, - &encoding_flags)) { + encoding_info = ext4_sb_read_encoding(es); + if (!encoding_info) { ext4_msg(sb, KERN_ERR, "Encoding requested by superblock is unknown"); goto failed_mount; @@ -4636,15 +4626,21 @@ static int __ext4_fill_super(struct fs_context *fc, struct super_block *sb) encoding = utf8_load(encoding_info->version); if (IS_ERR(encoding)) { ext4_msg(sb, KERN_ERR, - "can't mount with superblock charset: %s-%s " + "can't mount with superblock charset: %s-%u.%u.%u " "not supported by the kernel. flags: 0x%x.", - encoding_info->name, encoding_info->version, + encoding_info->name, + unicode_major(encoding_info->version), + unicode_minor(encoding_info->version), + unicode_rev(encoding_info->version), encoding_flags); goto failed_mount; } ext4_msg(sb, KERN_INFO,"Using encoding defined by superblock: " - "%s-%s with flags 0x%hx", encoding_info->name, - encoding_info->version?:"\b", encoding_flags); + "%s-%u.%u.%u with flags 0x%hx", encoding_info->name, + unicode_major(encoding_info->version), + unicode_minor(encoding_info->version), + unicode_rev(encoding_info->version), + encoding_flags); sb->s_encoding = encoding; sb->s_encoding_flags = encoding_flags; diff --git a/fs/f2fs/Kconfig b/fs/f2fs/Kconfig index 7eea3cfd894d..f46a7339d6cf 100644 --- a/fs/f2fs/Kconfig +++ b/fs/f2fs/Kconfig @@ -7,6 +7,7 @@ config F2FS_FS select CRYPTO_CRC32 select F2FS_FS_XATTR if FS_ENCRYPTION select FS_ENCRYPTION_ALGS if FS_ENCRYPTION + select FS_IOMAP select LZ4_COMPRESS if F2FS_FS_LZ4 select LZ4_DECOMPRESS if F2FS_FS_LZ4 select LZ4HC_COMPRESS if F2FS_FS_LZ4HC diff --git a/fs/f2fs/checkpoint.c b/fs/f2fs/checkpoint.c index f1693d45bb78..982f0170639f 100644 --- a/fs/f2fs/checkpoint.c +++ b/fs/f2fs/checkpoint.c @@ -664,7 +664,7 @@ static int recover_orphan_inode(struct f2fs_sb_info *sbi, nid_t ino) /* truncate all the data during iput */ iput(inode); - err = f2fs_get_node_info(sbi, ino, &ni); + err = f2fs_get_node_info(sbi, ino, &ni, false); if (err) goto err_out; @@ -1302,8 +1302,8 @@ static void update_ckpt_flags(struct f2fs_sb_info *sbi, struct cp_control *cpc) unsigned long flags; if (cpc->reason & CP_UMOUNT) { - if (le32_to_cpu(ckpt->cp_pack_total_block_count) > - sbi->blocks_per_seg - NM_I(sbi)->nat_bits_blocks) { + if (le32_to_cpu(ckpt->cp_pack_total_block_count) + + NM_I(sbi)->nat_bits_blocks > sbi->blocks_per_seg) { clear_ckpt_flags(sbi, CP_NAT_BITS_FLAG); f2fs_notice(sbi, "Disable nat_bits due to no space"); } else if (!is_set_ckpt_flags(sbi, CP_NAT_BITS_FLAG) && diff --git a/fs/f2fs/compress.c b/fs/f2fs/compress.c index 49121a21f749..d0c3aeba5945 100644 --- a/fs/f2fs/compress.c +++ b/fs/f2fs/compress.c @@ -154,6 +154,7 @@ void f2fs_destroy_compress_ctx(struct compress_ctx *cc, bool reuse) cc->rpages = NULL; cc->nr_rpages = 0; cc->nr_cpages = 0; + cc->valid_nr_cpages = 0; if (!reuse) cc->cluster_idx = NULL_CLUSTER; } @@ -620,7 +621,6 @@ static int f2fs_compress_pages(struct compress_ctx *cc) const struct f2fs_compress_ops *cops = f2fs_cops[fi->i_compress_algorithm]; unsigned int max_len, new_nr_cpages; - struct page **new_cpages; u32 chksum = 0; int i, ret; @@ -635,6 +635,7 @@ static int f2fs_compress_pages(struct compress_ctx *cc) max_len = COMPRESS_HEADER_SIZE + cc->clen; cc->nr_cpages = DIV_ROUND_UP(max_len, PAGE_SIZE); + cc->valid_nr_cpages = cc->nr_cpages; cc->cpages = page_array_alloc(cc->inode, cc->nr_cpages); if (!cc->cpages) { @@ -685,13 +686,6 @@ static int f2fs_compress_pages(struct compress_ctx *cc) new_nr_cpages = DIV_ROUND_UP(cc->clen + COMPRESS_HEADER_SIZE, PAGE_SIZE); - /* Now we're going to cut unnecessary tail pages */ - new_cpages = page_array_alloc(cc->inode, new_nr_cpages); - if (!new_cpages) { - ret = -ENOMEM; - goto out_vunmap_cbuf; - } - /* zero out any unused part of the last page */ memset(&cc->cbuf->cdata[cc->clen], 0, (new_nr_cpages * PAGE_SIZE) - @@ -701,10 +695,8 @@ static int f2fs_compress_pages(struct compress_ctx *cc) vm_unmap_ram(cc->rbuf, cc->cluster_size); for (i = 0; i < cc->nr_cpages; i++) { - if (i < new_nr_cpages) { - new_cpages[i] = cc->cpages[i]; + if (i < new_nr_cpages) continue; - } f2fs_compress_free_page(cc->cpages[i]); cc->cpages[i] = NULL; } @@ -712,9 +704,7 @@ static int f2fs_compress_pages(struct compress_ctx *cc) if (cops->destroy_compress_ctx) cops->destroy_compress_ctx(cc); - page_array_free(cc->inode, cc->cpages, cc->nr_cpages); - cc->cpages = new_cpages; - cc->nr_cpages = new_nr_cpages; + cc->valid_nr_cpages = new_nr_cpages; trace_f2fs_compress_pages_end(cc->inode, cc->cluster_idx, cc->clen, ret); @@ -1296,7 +1286,7 @@ static int f2fs_write_compressed_pages(struct compress_ctx *cc, psize = (loff_t)(cc->rpages[last_index]->index + 1) << PAGE_SHIFT; - err = f2fs_get_node_info(fio.sbi, dn.nid, &ni); + err = f2fs_get_node_info(fio.sbi, dn.nid, &ni, false); if (err) goto out_put_dnode; @@ -1308,14 +1298,14 @@ static int f2fs_write_compressed_pages(struct compress_ctx *cc, cic->magic = F2FS_COMPRESSED_PAGE_MAGIC; cic->inode = inode; - atomic_set(&cic->pending_pages, cc->nr_cpages); + atomic_set(&cic->pending_pages, cc->valid_nr_cpages); cic->rpages = page_array_alloc(cc->inode, cc->cluster_size); if (!cic->rpages) goto out_put_cic; cic->nr_rpages = cc->cluster_size; - for (i = 0; i < cc->nr_cpages; i++) { + for (i = 0; i < cc->valid_nr_cpages; i++) { f2fs_set_compressed_page(cc->cpages[i], inode, cc->rpages[i + 1]->index, cic); fio.compressed_page = cc->cpages[i]; @@ -1360,7 +1350,7 @@ static int f2fs_write_compressed_pages(struct compress_ctx *cc, if (fio.compr_blocks && __is_valid_data_blkaddr(blkaddr)) fio.compr_blocks++; - if (i > cc->nr_cpages) { + if (i > cc->valid_nr_cpages) { if (__is_valid_data_blkaddr(blkaddr)) { f2fs_invalidate_blocks(sbi, blkaddr); f2fs_update_data_blkaddr(&dn, NEW_ADDR); @@ -1385,8 +1375,8 @@ unlock_continue: if (fio.compr_blocks) f2fs_i_compr_blocks_update(inode, fio.compr_blocks - 1, false); - f2fs_i_compr_blocks_update(inode, cc->nr_cpages, true); - add_compr_block_stat(inode, cc->nr_cpages); + f2fs_i_compr_blocks_update(inode, cc->valid_nr_cpages, true); + add_compr_block_stat(inode, cc->valid_nr_cpages); set_inode_flag(cc->inode, FI_APPEND_WRITE); if (cc->cluster_idx == 0) @@ -1424,9 +1414,7 @@ out_unlock_op: else f2fs_unlock_op(sbi); out_free: - for (i = 0; i < cc->nr_cpages; i++) { - if (!cc->cpages[i]) - continue; + for (i = 0; i < cc->valid_nr_cpages; i++) { f2fs_compress_free_page(cc->cpages[i]); cc->cpages[i] = NULL; } @@ -1468,25 +1456,38 @@ static int f2fs_write_raw_pages(struct compress_ctx *cc, enum iostat_type io_type) { struct address_space *mapping = cc->inode->i_mapping; - int _submitted, compr_blocks, ret; - int i = -1, err = 0; + int _submitted, compr_blocks, ret, i; compr_blocks = f2fs_compressed_blocks(cc); - if (compr_blocks < 0) { - err = compr_blocks; - goto out_err; + + for (i = 0; i < cc->cluster_size; i++) { + if (!cc->rpages[i]) + continue; + + redirty_page_for_writepage(wbc, cc->rpages[i]); + unlock_page(cc->rpages[i]); } + if (compr_blocks < 0) + return compr_blocks; + for (i = 0; i < cc->cluster_size; i++) { if (!cc->rpages[i]) continue; retry_write: + lock_page(cc->rpages[i]); + if (cc->rpages[i]->mapping != mapping) { +continue_unlock: unlock_page(cc->rpages[i]); continue; } - BUG_ON(!PageLocked(cc->rpages[i])); + if (!PageDirty(cc->rpages[i])) + goto continue_unlock; + + if (!clear_page_dirty_for_io(cc->rpages[i])) + goto continue_unlock; ret = f2fs_write_single_data_page(cc->rpages[i], &_submitted, NULL, NULL, wbc, io_type, @@ -1501,26 +1502,15 @@ retry_write: * avoid deadlock caused by cluster update race * from foreground operation. */ - if (IS_NOQUOTA(cc->inode)) { - err = 0; - goto out_err; - } + if (IS_NOQUOTA(cc->inode)) + return 0; ret = 0; cond_resched(); congestion_wait(BLK_RW_ASYNC, DEFAULT_IO_TIMEOUT); - lock_page(cc->rpages[i]); - - if (!PageDirty(cc->rpages[i])) { - unlock_page(cc->rpages[i]); - continue; - } - - clear_page_dirty_for_io(cc->rpages[i]); goto retry_write; } - err = ret; - goto out_err; + return ret; } *submitted += _submitted; @@ -1529,14 +1519,6 @@ retry_write: f2fs_balance_fs(F2FS_M_SB(mapping), true); return 0; -out_err: - for (++i; i < cc->cluster_size; i++) { - if (!cc->rpages[i]) - continue; - redirty_page_for_writepage(wbc, cc->rpages[i]); - unlock_page(cc->rpages[i]); - } - return err; } int f2fs_write_multi_pages(struct compress_ctx *cc, diff --git a/fs/f2fs/data.c b/fs/f2fs/data.c index 9f754aaef558..8c417864c66a 100644 --- a/fs/f2fs/data.c +++ b/fs/f2fs/data.c @@ -8,9 +8,9 @@ #include <linux/fs.h> #include <linux/f2fs_fs.h> #include <linux/buffer_head.h> +#include <linux/sched/mm.h> #include <linux/mpage.h> #include <linux/writeback.h> -#include <linux/backing-dev.h> #include <linux/pagevec.h> #include <linux/blkdev.h> #include <linux/bio.h> @@ -18,9 +18,9 @@ #include <linux/swap.h> #include <linux/prefetch.h> #include <linux/uio.h> -#include <linux/cleancache.h> #include <linux/sched/signal.h> #include <linux/fiemap.h> +#include <linux/iomap.h> #include "f2fs.h" #include "node.h" @@ -1354,7 +1354,7 @@ static int __allocate_data_block(struct dnode_of_data *dn, int seg_type) if (unlikely(is_inode_flag_set(dn->inode, FI_NO_ALLOC))) return -EPERM; - err = f2fs_get_node_info(sbi, dn->nid, &ni); + err = f2fs_get_node_info(sbi, dn->nid, &ni, false); if (err) return err; @@ -1376,61 +1376,9 @@ alloc: f2fs_invalidate_compress_page(sbi, old_blkaddr); } f2fs_update_data_blkaddr(dn, dn->data_blkaddr); - - /* - * i_size will be updated by direct_IO. Otherwise, we'll get stale - * data from unwritten block via dio_read. - */ return 0; } -int f2fs_preallocate_blocks(struct kiocb *iocb, struct iov_iter *from) -{ - struct inode *inode = file_inode(iocb->ki_filp); - struct f2fs_map_blocks map; - int flag; - int err = 0; - bool direct_io = iocb->ki_flags & IOCB_DIRECT; - - map.m_lblk = F2FS_BLK_ALIGN(iocb->ki_pos); - map.m_len = F2FS_BYTES_TO_BLK(iocb->ki_pos + iov_iter_count(from)); - if (map.m_len > map.m_lblk) - map.m_len -= map.m_lblk; - else - map.m_len = 0; - - map.m_next_pgofs = NULL; - map.m_next_extent = NULL; - map.m_seg_type = NO_CHECK_TYPE; - map.m_may_create = true; - - if (direct_io) { - map.m_seg_type = f2fs_rw_hint_to_seg_type(iocb->ki_hint); - flag = f2fs_force_buffered_io(inode, iocb, from) ? - F2FS_GET_BLOCK_PRE_AIO : - F2FS_GET_BLOCK_PRE_DIO; - goto map_blocks; - } - if (iocb->ki_pos + iov_iter_count(from) > MAX_INLINE_DATA(inode)) { - err = f2fs_convert_inline_inode(inode); - if (err) - return err; - } - if (f2fs_has_inline_data(inode)) - return err; - - flag = F2FS_GET_BLOCK_PRE_AIO; - -map_blocks: - err = f2fs_map_blocks(inode, &map, 1, flag); - if (map.m_len > 0 && err == -ENOSPC) { - if (!direct_io) - set_inode_flag(inode, FI_NO_PREALLOC); - err = 0; - } - return err; -} - void f2fs_do_map_lock(struct f2fs_sb_info *sbi, int flag, bool lock) { if (flag == F2FS_GET_BLOCK_PRE_AIO) { @@ -1590,8 +1538,11 @@ next_block: flag != F2FS_GET_BLOCK_DIO); err = __allocate_data_block(&dn, map->m_seg_type); - if (!err) + if (!err) { + if (flag == F2FS_GET_BLOCK_PRE_DIO) + file_need_truncate(inode); set_inode_flag(inode, FI_APPEND_WRITE); + } } if (err) goto sync_out; @@ -1786,50 +1737,6 @@ static inline u64 blks_to_bytes(struct inode *inode, u64 blks) return (blks << inode->i_blkbits); } -static int __get_data_block(struct inode *inode, sector_t iblock, - struct buffer_head *bh, int create, int flag, - pgoff_t *next_pgofs, int seg_type, bool may_write) -{ - struct f2fs_map_blocks map; - int err; - - map.m_lblk = iblock; - map.m_len = bytes_to_blks(inode, bh->b_size); - map.m_next_pgofs = next_pgofs; - map.m_next_extent = NULL; - map.m_seg_type = seg_type; - map.m_may_create = may_write; - - err = f2fs_map_blocks(inode, &map, create, flag); - if (!err) { - map_bh(bh, inode->i_sb, map.m_pblk); - bh->b_state = (bh->b_state & ~F2FS_MAP_FLAGS) | map.m_flags; - bh->b_size = blks_to_bytes(inode, map.m_len); - - if (map.m_multidev_dio) - bh->b_bdev = map.m_bdev; - } - return err; -} - -static int get_data_block_dio_write(struct inode *inode, sector_t iblock, - struct buffer_head *bh_result, int create) -{ - return __get_data_block(inode, iblock, bh_result, create, - F2FS_GET_BLOCK_DIO, NULL, - f2fs_rw_hint_to_seg_type(inode->i_write_hint), - true); -} - -static int get_data_block_dio(struct inode *inode, sector_t iblock, - struct buffer_head *bh_result, int create) -{ - return __get_data_block(inode, iblock, bh_result, create, - F2FS_GET_BLOCK_DIO, NULL, - f2fs_rw_hint_to_seg_type(inode->i_write_hint), - false); -} - static int f2fs_xattr_fiemap(struct inode *inode, struct fiemap_extent_info *fieinfo) { @@ -1849,7 +1756,7 @@ static int f2fs_xattr_fiemap(struct inode *inode, if (!page) return -ENOMEM; - err = f2fs_get_node_info(sbi, inode->i_ino, &ni); + err = f2fs_get_node_info(sbi, inode->i_ino, &ni, false); if (err) { f2fs_put_page(page, 1); return err; @@ -1881,7 +1788,7 @@ static int f2fs_xattr_fiemap(struct inode *inode, if (!page) return -ENOMEM; - err = f2fs_get_node_info(sbi, xnid, &ni); + err = f2fs_get_node_info(sbi, xnid, &ni, false); if (err) { f2fs_put_page(page, 1); return err; @@ -2127,12 +2034,6 @@ got_it: block_nr = map->m_pblk + block_in_file - map->m_lblk; SetPageMappedToDisk(page); - if (!PageUptodate(page) && (!PageSwapCache(page) && - !cleancache_get_page(page))) { - SetPageUptodate(page); - goto confused; - } - if (!f2fs_is_valid_blkaddr(F2FS_I_SB(inode), block_nr, DATA_GENERIC_ENHANCE_READ)) { ret = -EFSCORRUPTED; @@ -2188,12 +2089,6 @@ submit_and_realloc: ClearPageError(page); *last_block_in_bio = block_nr; goto out; -confused: - if (bio) { - __submit_bio(F2FS_I_SB(inode), bio, DATA); - bio = NULL; - } - unlock_page(page); out: *bio_ret = bio; return ret; @@ -2542,7 +2437,7 @@ retry_encrypt: /* flush pending IOs and wait for a while in the ENOMEM case */ if (PTR_ERR(fio->encrypted_page) == -ENOMEM) { f2fs_flush_merged_writes(fio->sbi); - congestion_wait(BLK_RW_ASYNC, DEFAULT_IO_TIMEOUT); + memalloc_retry_wait(GFP_NOFS); gfp_flags |= __GFP_NOFAIL; goto retry_encrypt; } @@ -2617,6 +2512,11 @@ bool f2fs_should_update_outplace(struct inode *inode, struct f2fs_io_info *fio) { struct f2fs_sb_info *sbi = F2FS_I_SB(inode); + /* The below cases were checked when setting it. */ + if (f2fs_is_pinned_file(inode)) + return false; + if (fio && is_sbi_flag_set(sbi, SBI_NEED_FSCK)) + return true; if (f2fs_lfs_mode(sbi)) return true; if (S_ISDIR(inode->i_mode)) @@ -2625,8 +2525,6 @@ bool f2fs_should_update_outplace(struct inode *inode, struct f2fs_io_info *fio) return true; if (f2fs_is_atomic_file(inode)) return true; - if (is_sbi_flag_set(sbi, SBI_NEED_FSCK)) - return true; /* swap file is migrating in aligned write mode */ if (is_inode_flag_set(inode, FI_ALIGNED_WRITE)) @@ -2738,7 +2636,7 @@ got_it: fio->need_lock = LOCK_REQ; } - err = f2fs_get_node_info(fio->sbi, dn.nid, &ni); + err = f2fs_get_node_info(fio->sbi, dn.nid, &ni, false); if (err) goto out_writepage; @@ -2987,6 +2885,7 @@ static int f2fs_write_cache_pages(struct address_space *mapping, .rpages = NULL, .nr_rpages = 0, .cpages = NULL, + .valid_nr_cpages = 0, .rbuf = NULL, .cbuf = NULL, .rlen = PAGE_SIZE * F2FS_I(inode)->i_cluster_size, @@ -3305,7 +3204,7 @@ static int f2fs_write_data_pages(struct address_space *mapping, FS_CP_DATA_IO : FS_DATA_IO); } -static void f2fs_write_failed(struct inode *inode, loff_t to) +void f2fs_write_failed(struct inode *inode, loff_t to) { loff_t i_size = i_size_read(inode); @@ -3339,12 +3238,10 @@ static int prepare_write_begin(struct f2fs_sb_info *sbi, int flag; /* - * we already allocated all the blocks, so we don't need to get - * the block addresses when there is no need to fill the page. + * If a whole page is being written and we already preallocated all the + * blocks, then there is no need to get a block address now. */ - if (!f2fs_has_inline_data(inode) && len == PAGE_SIZE && - !is_inode_flag_set(inode, FI_NO_PREALLOC) && - !f2fs_verity_in_progress(inode)) + if (len == PAGE_SIZE && is_inode_flag_set(inode, FI_PREALLOCATED_ALL)) return 0; /* f2fs_lock_op avoids race between write CP and convert_inline_page */ @@ -3595,158 +3492,6 @@ unlock_out: return copied; } -static int check_direct_IO(struct inode *inode, struct iov_iter *iter, - loff_t offset) -{ - unsigned i_blkbits = READ_ONCE(inode->i_blkbits); - unsigned blkbits = i_blkbits; - unsigned blocksize_mask = (1 << blkbits) - 1; - unsigned long align = offset | iov_iter_alignment(iter); - struct block_device *bdev = inode->i_sb->s_bdev; - - if (iov_iter_rw(iter) == READ && offset >= i_size_read(inode)) - return 1; - - if (align & blocksize_mask) { - if (bdev) - blkbits = blksize_bits(bdev_logical_block_size(bdev)); - blocksize_mask = (1 << blkbits) - 1; - if (align & blocksize_mask) - return -EINVAL; - return 1; - } - return 0; -} - -static void f2fs_dio_end_io(struct bio *bio) -{ - struct f2fs_private_dio *dio = bio->bi_private; - - dec_page_count(F2FS_I_SB(dio->inode), - dio->write ? F2FS_DIO_WRITE : F2FS_DIO_READ); - - bio->bi_private = dio->orig_private; - bio->bi_end_io = dio->orig_end_io; - - kfree(dio); - - bio_endio(bio); -} - -static void f2fs_dio_submit_bio(struct bio *bio, struct inode *inode, - loff_t file_offset) -{ - struct f2fs_private_dio *dio; - bool write = (bio_op(bio) == REQ_OP_WRITE); - - dio = f2fs_kzalloc(F2FS_I_SB(inode), - sizeof(struct f2fs_private_dio), GFP_NOFS); - if (!dio) - goto out; - - dio->inode = inode; - dio->orig_end_io = bio->bi_end_io; - dio->orig_private = bio->bi_private; - dio->write = write; - - bio->bi_end_io = f2fs_dio_end_io; - bio->bi_private = dio; - - inc_page_count(F2FS_I_SB(inode), - write ? F2FS_DIO_WRITE : F2FS_DIO_READ); - - submit_bio(bio); - return; -out: - bio->bi_status = BLK_STS_IOERR; - bio_endio(bio); -} - -static ssize_t f2fs_direct_IO(struct kiocb *iocb, struct iov_iter *iter) -{ - struct address_space *mapping = iocb->ki_filp->f_mapping; - struct inode *inode = mapping->host; - struct f2fs_sb_info *sbi = F2FS_I_SB(inode); - struct f2fs_inode_info *fi = F2FS_I(inode); - size_t count = iov_iter_count(iter); - loff_t offset = iocb->ki_pos; - int rw = iov_iter_rw(iter); - int err; - enum rw_hint hint = iocb->ki_hint; - int whint_mode = F2FS_OPTION(sbi).whint_mode; - bool do_opu; - - err = check_direct_IO(inode, iter, offset); - if (err) - return err < 0 ? err : 0; - - if (f2fs_force_buffered_io(inode, iocb, iter)) - return 0; - - do_opu = rw == WRITE && f2fs_lfs_mode(sbi); - - trace_f2fs_direct_IO_enter(inode, offset, count, rw); - - if (rw == WRITE && whint_mode == WHINT_MODE_OFF) - iocb->ki_hint = WRITE_LIFE_NOT_SET; - - if (iocb->ki_flags & IOCB_NOWAIT) { - if (!down_read_trylock(&fi->i_gc_rwsem[rw])) { - iocb->ki_hint = hint; - err = -EAGAIN; - goto out; - } - if (do_opu && !down_read_trylock(&fi->i_gc_rwsem[READ])) { - up_read(&fi->i_gc_rwsem[rw]); - iocb->ki_hint = hint; - err = -EAGAIN; - goto out; - } - } else { - down_read(&fi->i_gc_rwsem[rw]); - if (do_opu) - down_read(&fi->i_gc_rwsem[READ]); - } - - err = __blockdev_direct_IO(iocb, inode, inode->i_sb->s_bdev, - iter, rw == WRITE ? get_data_block_dio_write : - get_data_block_dio, NULL, f2fs_dio_submit_bio, - rw == WRITE ? DIO_LOCKING | DIO_SKIP_HOLES : - DIO_SKIP_HOLES); - - if (do_opu) - up_read(&fi->i_gc_rwsem[READ]); - - up_read(&fi->i_gc_rwsem[rw]); - - if (rw == WRITE) { - if (whint_mode == WHINT_MODE_OFF) - iocb->ki_hint = hint; - if (err > 0) { - f2fs_update_iostat(F2FS_I_SB(inode), APP_DIRECT_IO, - err); - if (!do_opu) - set_inode_flag(inode, FI_UPDATE_WRITE); - } else if (err == -EIOCBQUEUED) { - f2fs_update_iostat(F2FS_I_SB(inode), APP_DIRECT_IO, - count - iov_iter_count(iter)); - } else if (err < 0) { - f2fs_write_failed(inode, offset + count); - } - } else { - if (err > 0) - f2fs_update_iostat(sbi, APP_DIRECT_READ_IO, err); - else if (err == -EIOCBQUEUED) - f2fs_update_iostat(F2FS_I_SB(inode), APP_DIRECT_READ_IO, - count - iov_iter_count(iter)); - } - -out: - trace_f2fs_direct_IO_exit(inode, offset, count, rw, err); - - return err; -} - void f2fs_invalidate_page(struct page *page, unsigned int offset, unsigned int length) { @@ -3770,12 +3515,9 @@ void f2fs_invalidate_page(struct page *page, unsigned int offset, clear_page_private_gcing(page); - if (test_opt(sbi, COMPRESS_CACHE)) { - if (f2fs_compressed_file(inode)) - f2fs_invalidate_compress_pages(sbi, inode->i_ino); - if (inode->i_ino == F2FS_COMPRESS_INO(sbi)) - clear_page_private_data(page); - } + if (test_opt(sbi, COMPRESS_CACHE) && + inode->i_ino == F2FS_COMPRESS_INO(sbi)) + clear_page_private_data(page); if (page_private_atomic(page)) return f2fs_drop_inmem_page(inode, page); @@ -3795,12 +3537,9 @@ int f2fs_release_page(struct page *page, gfp_t wait) return 0; if (test_opt(F2FS_P_SB(page), COMPRESS_CACHE)) { - struct f2fs_sb_info *sbi = F2FS_P_SB(page); struct inode *inode = page->mapping->host; - if (f2fs_compressed_file(inode)) - f2fs_invalidate_compress_pages(sbi, inode->i_ino); - if (inode->i_ino == F2FS_COMPRESS_INO(sbi)) + if (inode->i_ino == F2FS_COMPRESS_INO(F2FS_I_SB(inode))) clear_page_private_data(page); } @@ -4202,7 +3941,7 @@ const struct address_space_operations f2fs_dblock_aops = { .set_page_dirty = f2fs_set_data_page_dirty, .invalidatepage = f2fs_invalidate_page, .releasepage = f2fs_release_page, - .direct_IO = f2fs_direct_IO, + .direct_IO = noop_direct_IO, .bmap = f2fs_bmap, .swap_activate = f2fs_swap_activate, .swap_deactivate = f2fs_swap_deactivate, @@ -4282,3 +4021,58 @@ void f2fs_destroy_bio_entry_cache(void) { kmem_cache_destroy(bio_entry_slab); } + +static int f2fs_iomap_begin(struct inode *inode, loff_t offset, loff_t length, + unsigned int flags, struct iomap *iomap, + struct iomap *srcmap) +{ + struct f2fs_map_blocks map = {}; + pgoff_t next_pgofs = 0; + int err; + + map.m_lblk = bytes_to_blks(inode, offset); + map.m_len = bytes_to_blks(inode, offset + length - 1) - map.m_lblk + 1; + map.m_next_pgofs = &next_pgofs; + map.m_seg_type = f2fs_rw_hint_to_seg_type(inode->i_write_hint); + if (flags & IOMAP_WRITE) + map.m_may_create = true; + + err = f2fs_map_blocks(inode, &map, flags & IOMAP_WRITE, + F2FS_GET_BLOCK_DIO); + if (err) + return err; + + iomap->offset = blks_to_bytes(inode, map.m_lblk); + + if (map.m_flags & (F2FS_MAP_MAPPED | F2FS_MAP_UNWRITTEN)) { + iomap->length = blks_to_bytes(inode, map.m_len); + if (map.m_flags & F2FS_MAP_MAPPED) { + iomap->type = IOMAP_MAPPED; + iomap->flags |= IOMAP_F_MERGED; + } else { + iomap->type = IOMAP_UNWRITTEN; + } + if (WARN_ON_ONCE(!__is_valid_data_blkaddr(map.m_pblk))) + return -EINVAL; + + iomap->bdev = map.m_bdev; + iomap->addr = blks_to_bytes(inode, map.m_pblk); + } else { + iomap->length = blks_to_bytes(inode, next_pgofs) - + iomap->offset; + iomap->type = IOMAP_HOLE; + iomap->addr = IOMAP_NULL_ADDR; + } + + if (map.m_flags & F2FS_MAP_NEW) + iomap->flags |= IOMAP_F_NEW; + if ((inode->i_state & I_DIRTY_DATASYNC) || + offset + length > i_size_read(inode)) + iomap->flags |= IOMAP_F_DIRTY; + + return 0; +} + +const struct iomap_ops f2fs_iomap_ops = { + .iomap_begin = f2fs_iomap_begin, +}; diff --git a/fs/f2fs/f2fs.h b/fs/f2fs/f2fs.h index d0d603187171..eb22fa91c2b2 100644 --- a/fs/f2fs/f2fs.h +++ b/fs/f2fs/f2fs.h @@ -58,6 +58,7 @@ enum { FAULT_WRITE_IO, FAULT_SLAB_ALLOC, FAULT_DQUOT_INIT, + FAULT_LOCK_OP, FAULT_MAX, }; @@ -656,6 +657,7 @@ enum { #define FADVISE_KEEP_SIZE_BIT 0x10 #define FADVISE_HOT_BIT 0x20 #define FADVISE_VERITY_BIT 0x40 +#define FADVISE_TRUNC_BIT 0x80 #define FADVISE_MODIFIABLE_BITS (FADVISE_COLD_BIT | FADVISE_HOT_BIT) @@ -683,6 +685,10 @@ enum { #define file_is_verity(inode) is_file(inode, FADVISE_VERITY_BIT) #define file_set_verity(inode) set_file(inode, FADVISE_VERITY_BIT) +#define file_should_truncate(inode) is_file(inode, FADVISE_TRUNC_BIT) +#define file_need_truncate(inode) set_file(inode, FADVISE_TRUNC_BIT) +#define file_dont_truncate(inode) clear_file(inode, FADVISE_TRUNC_BIT) + #define DEF_DIR_LEVEL 0 enum { @@ -717,7 +723,7 @@ enum { FI_INLINE_DOTS, /* indicate inline dot dentries */ FI_DO_DEFRAG, /* indicate defragment is running */ FI_DIRTY_FILE, /* indicate regular/symlink has dirty pages */ - FI_NO_PREALLOC, /* indicate skipped preallocated blocks */ + FI_PREALLOCATED_ALL, /* all blocks for write were preallocated */ FI_HOT_DATA, /* indicate file is hot */ FI_EXTRA_ATTR, /* indicate file has extra attribute */ FI_PROJ_INHERIT, /* indicate file inherits projectid */ @@ -1020,6 +1026,7 @@ struct f2fs_sm_info { unsigned int segment_count; /* total # of segments */ unsigned int main_segments; /* # of segments in main area */ unsigned int reserved_segments; /* # of reserved segments */ + unsigned int additional_reserved_segments;/* reserved segs for IO align feature */ unsigned int ovp_segments; /* # of overprovision segments */ /* a threshold to reclaim prefree segments */ @@ -1488,6 +1495,7 @@ struct compress_ctx { unsigned int nr_rpages; /* total page number in rpages */ struct page **cpages; /* pages store compressed data in cluster */ unsigned int nr_cpages; /* total page number in cpages */ + unsigned int valid_nr_cpages; /* valid page number in cpages */ void *rbuf; /* virtual mapped address on rpages */ struct compress_data *cbuf; /* virtual mapped address on cpages */ size_t rlen; /* valid data length in rbuf */ @@ -1679,6 +1687,9 @@ struct f2fs_sb_info { unsigned int cur_victim_sec; /* current victim section num */ unsigned int gc_mode; /* current GC state */ unsigned int next_victim_seg[2]; /* next segment in victim section */ + spinlock_t gc_urgent_high_lock; + bool gc_urgent_high_limited; /* indicates having limited trial count */ + unsigned int gc_urgent_high_remaining; /* remaining trial count for GC_URGENT_HIGH */ /* for skip statistic */ unsigned int atomic_files; /* # of opened atomic file */ @@ -1803,13 +1814,6 @@ struct f2fs_sb_info { #endif }; -struct f2fs_private_dio { - struct inode *inode; - void *orig_private; - bio_end_io_t *orig_end_io; - bool write; -}; - #ifdef CONFIG_F2FS_FAULT_INJECTION #define f2fs_show_injection_info(sbi, type) \ printk_ratelimited("%sF2FS-fs (%s) : inject %s in %s of %pS\n", \ @@ -2095,6 +2099,10 @@ static inline void f2fs_lock_op(struct f2fs_sb_info *sbi) static inline int f2fs_trylock_op(struct f2fs_sb_info *sbi) { + if (time_to_inject(sbi, FAULT_LOCK_OP)) { + f2fs_show_injection_info(sbi, FAULT_LOCK_OP); + return 0; + } return down_read_trylock(&sbi->cp_rwsem); } @@ -2200,6 +2208,11 @@ static inline int inc_valid_block_count(struct f2fs_sb_info *sbi, if (!__allow_reserved_blocks(sbi, inode, true)) avail_user_block_count -= F2FS_OPTION(sbi).root_reserved_blocks; + + if (F2FS_IO_ALIGNED(sbi)) + avail_user_block_count -= sbi->blocks_per_seg * + SM_I(sbi)->additional_reserved_segments; + if (unlikely(is_sbi_flag_set(sbi, SBI_CP_DISABLED))) { if (avail_user_block_count > sbi->unusable_block_count) avail_user_block_count -= sbi->unusable_block_count; @@ -2446,6 +2459,11 @@ static inline int inc_valid_node_count(struct f2fs_sb_info *sbi, if (!__allow_reserved_blocks(sbi, inode, false)) valid_block_count += F2FS_OPTION(sbi).root_reserved_blocks; + + if (F2FS_IO_ALIGNED(sbi)) + valid_block_count += sbi->blocks_per_seg * + SM_I(sbi)->additional_reserved_segments; + user_block_count = sbi->user_block_count; if (unlikely(is_sbi_flag_set(sbi, SBI_CP_DISABLED))) user_block_count -= sbi->unusable_block_count; @@ -3118,12 +3136,16 @@ static inline int is_file(struct inode *inode, int type) static inline void set_file(struct inode *inode, int type) { + if (is_file(inode, type)) + return; F2FS_I(inode)->i_advise |= type; f2fs_mark_inode_dirty_sync(inode, true); } static inline void clear_file(struct inode *inode, int type) { + if (!is_file(inode, type)) + return; F2FS_I(inode)->i_advise &= ~type; f2fs_mark_inode_dirty_sync(inode, true); } @@ -3408,7 +3430,7 @@ int f2fs_need_dentry_mark(struct f2fs_sb_info *sbi, nid_t nid); bool f2fs_is_checkpointed_node(struct f2fs_sb_info *sbi, nid_t nid); bool f2fs_need_inode_block_update(struct f2fs_sb_info *sbi, nid_t ino); int f2fs_get_node_info(struct f2fs_sb_info *sbi, nid_t nid, - struct node_info *ni); + struct node_info *ni, bool checkpoint_context); pgoff_t f2fs_get_next_page_offset(struct dnode_of_data *dn, pgoff_t pgofs); int f2fs_get_dnode_of_data(struct dnode_of_data *dn, pgoff_t index, int mode); int f2fs_truncate_inode_blocks(struct inode *inode, pgoff_t from); @@ -3616,7 +3638,6 @@ void f2fs_update_data_blkaddr(struct dnode_of_data *dn, block_t blkaddr); int f2fs_reserve_new_blocks(struct dnode_of_data *dn, blkcnt_t count); int f2fs_reserve_new_block(struct dnode_of_data *dn); int f2fs_get_block(struct dnode_of_data *dn, pgoff_t index); -int f2fs_preallocate_blocks(struct kiocb *iocb, struct iov_iter *from); int f2fs_reserve_block(struct dnode_of_data *dn, pgoff_t index); struct page *f2fs_get_read_data_page(struct inode *inode, pgoff_t index, int op_flags, bool for_write); @@ -3639,6 +3660,7 @@ int f2fs_write_single_data_page(struct page *page, int *submitted, struct writeback_control *wbc, enum iostat_type io_type, int compr_blocks, bool allow_balance); +void f2fs_write_failed(struct inode *inode, loff_t to); void f2fs_invalidate_page(struct page *page, unsigned int offset, unsigned int length); int f2fs_release_page(struct page *page, gfp_t wait); @@ -3652,6 +3674,7 @@ int f2fs_init_post_read_processing(void); void f2fs_destroy_post_read_processing(void); int f2fs_init_post_read_wq(struct f2fs_sb_info *sbi); void f2fs_destroy_post_read_wq(struct f2fs_sb_info *sbi); +extern const struct iomap_ops f2fs_iomap_ops; /* * gc.c diff --git a/fs/f2fs/file.c b/fs/f2fs/file.c index 92ec2699bc85..3c98ef6af97d 100644 --- a/fs/f2fs/file.c +++ b/fs/f2fs/file.c @@ -24,6 +24,7 @@ #include <linux/sched/signal.h> #include <linux/fileattr.h> #include <linux/fadvise.h> +#include <linux/iomap.h> #include "f2fs.h" #include "node.h" @@ -1232,7 +1233,7 @@ static int __clone_blkaddrs(struct inode *src_inode, struct inode *dst_inode, if (ret) return ret; - ret = f2fs_get_node_info(sbi, dn.nid, &ni); + ret = f2fs_get_node_info(sbi, dn.nid, &ni, false); if (ret) { f2fs_put_dnode(&dn); return ret; @@ -1687,6 +1688,7 @@ next_alloc: map.m_seg_type = CURSEG_COLD_DATA_PINNED; err = f2fs_map_blocks(inode, &map, 1, F2FS_GET_BLOCK_PRE_DIO); + file_dont_truncate(inode); up_write(&sbi->pin_sem); @@ -1748,7 +1750,11 @@ static long f2fs_fallocate(struct file *file, int mode, (mode & (FALLOC_FL_COLLAPSE_RANGE | FALLOC_FL_INSERT_RANGE))) return -EOPNOTSUPP; - if (f2fs_compressed_file(inode) && + /* + * Pinned file should not support partial trucation since the block + * can be used by applications. + */ + if ((f2fs_compressed_file(inode) || f2fs_is_pinned_file(inode)) && (mode & (FALLOC_FL_PUNCH_HOLE | FALLOC_FL_COLLAPSE_RANGE | FALLOC_FL_ZERO_RANGE | FALLOC_FL_INSERT_RANGE))) return -EOPNOTSUPP; @@ -3143,17 +3149,17 @@ static int f2fs_ioc_set_pin_file(struct file *filp, unsigned long arg) inode_lock(inode); - if (f2fs_should_update_outplace(inode, NULL)) { - ret = -EINVAL; - goto out; - } - if (!pin) { clear_inode_flag(inode, FI_PIN_FILE); f2fs_i_gc_failures_write(inode, 0); goto done; } + if (f2fs_should_update_outplace(inode, NULL)) { + ret = -EINVAL; + goto out; + } + if (f2fs_pin_file_control(inode, false)) { ret = -EAGAIN; goto out; @@ -4218,27 +4224,385 @@ long f2fs_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) return __f2fs_ioctl(filp, cmd, arg); } -static ssize_t f2fs_file_read_iter(struct kiocb *iocb, struct iov_iter *iter) +/* + * Return %true if the given read or write request should use direct I/O, or + * %false if it should use buffered I/O. + */ +static bool f2fs_should_use_dio(struct inode *inode, struct kiocb *iocb, + struct iov_iter *iter) +{ + unsigned int align; + + if (!(iocb->ki_flags & IOCB_DIRECT)) + return false; + + if (f2fs_force_buffered_io(inode, iocb, iter)) + return false; + + /* + * Direct I/O not aligned to the disk's logical_block_size will be + * attempted, but will fail with -EINVAL. + * + * f2fs additionally requires that direct I/O be aligned to the + * filesystem block size, which is often a stricter requirement. + * However, f2fs traditionally falls back to buffered I/O on requests + * that are logical_block_size-aligned but not fs-block aligned. + * + * The below logic implements this behavior. + */ + align = iocb->ki_pos | iov_iter_alignment(iter); + if (!IS_ALIGNED(align, i_blocksize(inode)) && + IS_ALIGNED(align, bdev_logical_block_size(inode->i_sb->s_bdev))) + return false; + + return true; +} + +static int f2fs_dio_read_end_io(struct kiocb *iocb, ssize_t size, int error, + unsigned int flags) +{ + struct f2fs_sb_info *sbi = F2FS_I_SB(file_inode(iocb->ki_filp)); + + dec_page_count(sbi, F2FS_DIO_READ); + if (error) + return error; + f2fs_update_iostat(sbi, APP_DIRECT_READ_IO, size); + return 0; +} + +static const struct iomap_dio_ops f2fs_iomap_dio_read_ops = { + .end_io = f2fs_dio_read_end_io, +}; + +static ssize_t f2fs_dio_read_iter(struct kiocb *iocb, struct iov_iter *to) { struct file *file = iocb->ki_filp; struct inode *inode = file_inode(file); - int ret; + struct f2fs_sb_info *sbi = F2FS_I_SB(inode); + struct f2fs_inode_info *fi = F2FS_I(inode); + const loff_t pos = iocb->ki_pos; + const size_t count = iov_iter_count(to); + struct iomap_dio *dio; + ssize_t ret; + + if (count == 0) + return 0; /* skip atime update */ + + trace_f2fs_direct_IO_enter(inode, iocb, count, READ); + + if (iocb->ki_flags & IOCB_NOWAIT) { + if (!down_read_trylock(&fi->i_gc_rwsem[READ])) { + ret = -EAGAIN; + goto out; + } + } else { + down_read(&fi->i_gc_rwsem[READ]); + } + + /* + * We have to use __iomap_dio_rw() and iomap_dio_complete() instead of + * the higher-level function iomap_dio_rw() in order to ensure that the + * F2FS_DIO_READ counter will be decremented correctly in all cases. + */ + inc_page_count(sbi, F2FS_DIO_READ); + dio = __iomap_dio_rw(iocb, to, &f2fs_iomap_ops, + &f2fs_iomap_dio_read_ops, 0, 0); + if (IS_ERR_OR_NULL(dio)) { + ret = PTR_ERR_OR_ZERO(dio); + if (ret != -EIOCBQUEUED) + dec_page_count(sbi, F2FS_DIO_READ); + } else { + ret = iomap_dio_complete(dio); + } + + up_read(&fi->i_gc_rwsem[READ]); + + file_accessed(file); +out: + trace_f2fs_direct_IO_exit(inode, pos, count, READ, ret); + return ret; +} + +static ssize_t f2fs_file_read_iter(struct kiocb *iocb, struct iov_iter *to) +{ + struct inode *inode = file_inode(iocb->ki_filp); + ssize_t ret; if (!f2fs_is_compress_backend_ready(inode)) return -EOPNOTSUPP; - ret = generic_file_read_iter(iocb, iter); + if (f2fs_should_use_dio(inode, iocb, to)) + return f2fs_dio_read_iter(iocb, to); + ret = filemap_read(iocb, to, 0); if (ret > 0) - f2fs_update_iostat(F2FS_I_SB(inode), APP_READ_IO, ret); + f2fs_update_iostat(F2FS_I_SB(inode), APP_BUFFERED_READ_IO, ret); + return ret; +} + +static ssize_t f2fs_write_checks(struct kiocb *iocb, struct iov_iter *from) +{ + struct file *file = iocb->ki_filp; + struct inode *inode = file_inode(file); + ssize_t count; + int err; + + if (IS_IMMUTABLE(inode)) + return -EPERM; + + if (is_inode_flag_set(inode, FI_COMPRESS_RELEASED)) + return -EPERM; + + count = generic_write_checks(iocb, from); + if (count <= 0) + return count; + + err = file_modified(file); + if (err) + return err; + return count; +} + +/* + * Preallocate blocks for a write request, if it is possible and helpful to do + * so. Returns a positive number if blocks may have been preallocated, 0 if no + * blocks were preallocated, or a negative errno value if something went + * seriously wrong. Also sets FI_PREALLOCATED_ALL on the inode if *all* the + * requested blocks (not just some of them) have been allocated. + */ +static int f2fs_preallocate_blocks(struct kiocb *iocb, struct iov_iter *iter, + bool dio) +{ + struct inode *inode = file_inode(iocb->ki_filp); + struct f2fs_sb_info *sbi = F2FS_I_SB(inode); + const loff_t pos = iocb->ki_pos; + const size_t count = iov_iter_count(iter); + struct f2fs_map_blocks map = {}; + int flag; + int ret; + + /* If it will be an out-of-place direct write, don't bother. */ + if (dio && f2fs_lfs_mode(sbi)) + return 0; + /* + * Don't preallocate holes aligned to DIO_SKIP_HOLES which turns into + * buffered IO, if DIO meets any holes. + */ + if (dio && i_size_read(inode) && + (F2FS_BYTES_TO_BLK(pos) < F2FS_BLK_ALIGN(i_size_read(inode)))) + return 0; + + /* No-wait I/O can't allocate blocks. */ + if (iocb->ki_flags & IOCB_NOWAIT) + return 0; + + /* If it will be a short write, don't bother. */ + if (fault_in_iov_iter_readable(iter, count)) + return 0; + + if (f2fs_has_inline_data(inode)) { + /* If the data will fit inline, don't bother. */ + if (pos + count <= MAX_INLINE_DATA(inode)) + return 0; + ret = f2fs_convert_inline_inode(inode); + if (ret) + return ret; + } + + /* Do not preallocate blocks that will be written partially in 4KB. */ + map.m_lblk = F2FS_BLK_ALIGN(pos); + map.m_len = F2FS_BYTES_TO_BLK(pos + count); + if (map.m_len > map.m_lblk) + map.m_len -= map.m_lblk; + else + map.m_len = 0; + map.m_may_create = true; + if (dio) { + map.m_seg_type = f2fs_rw_hint_to_seg_type(inode->i_write_hint); + flag = F2FS_GET_BLOCK_PRE_DIO; + } else { + map.m_seg_type = NO_CHECK_TYPE; + flag = F2FS_GET_BLOCK_PRE_AIO; + } + + ret = f2fs_map_blocks(inode, &map, 1, flag); + /* -ENOSPC|-EDQUOT are fine to report the number of allocated blocks. */ + if (ret < 0 && !((ret == -ENOSPC || ret == -EDQUOT) && map.m_len > 0)) + return ret; + if (ret == 0) + set_inode_flag(inode, FI_PREALLOCATED_ALL); + return map.m_len; +} + +static ssize_t f2fs_buffered_write_iter(struct kiocb *iocb, + struct iov_iter *from) +{ + struct file *file = iocb->ki_filp; + struct inode *inode = file_inode(file); + ssize_t ret; + if (iocb->ki_flags & IOCB_NOWAIT) + return -EOPNOTSUPP; + + current->backing_dev_info = inode_to_bdi(inode); + ret = generic_perform_write(file, from, iocb->ki_pos); + current->backing_dev_info = NULL; + + if (ret > 0) { + iocb->ki_pos += ret; + f2fs_update_iostat(F2FS_I_SB(inode), APP_BUFFERED_IO, ret); + } return ret; } -static ssize_t f2fs_file_write_iter(struct kiocb *iocb, struct iov_iter *from) +static int f2fs_dio_write_end_io(struct kiocb *iocb, ssize_t size, int error, + unsigned int flags) +{ + struct f2fs_sb_info *sbi = F2FS_I_SB(file_inode(iocb->ki_filp)); + + dec_page_count(sbi, F2FS_DIO_WRITE); + if (error) + return error; + f2fs_update_iostat(sbi, APP_DIRECT_IO, size); + return 0; +} + +static const struct iomap_dio_ops f2fs_iomap_dio_write_ops = { + .end_io = f2fs_dio_write_end_io, +}; + +static ssize_t f2fs_dio_write_iter(struct kiocb *iocb, struct iov_iter *from, + bool *may_need_sync) { struct file *file = iocb->ki_filp; struct inode *inode = file_inode(file); + struct f2fs_inode_info *fi = F2FS_I(inode); + struct f2fs_sb_info *sbi = F2FS_I_SB(inode); + const bool do_opu = f2fs_lfs_mode(sbi); + const int whint_mode = F2FS_OPTION(sbi).whint_mode; + const loff_t pos = iocb->ki_pos; + const ssize_t count = iov_iter_count(from); + const enum rw_hint hint = iocb->ki_hint; + unsigned int dio_flags; + struct iomap_dio *dio; + ssize_t ret; + + trace_f2fs_direct_IO_enter(inode, iocb, count, WRITE); + + if (iocb->ki_flags & IOCB_NOWAIT) { + /* f2fs_convert_inline_inode() and block allocation can block */ + if (f2fs_has_inline_data(inode) || + !f2fs_overwrite_io(inode, pos, count)) { + ret = -EAGAIN; + goto out; + } + + if (!down_read_trylock(&fi->i_gc_rwsem[WRITE])) { + ret = -EAGAIN; + goto out; + } + if (do_opu && !down_read_trylock(&fi->i_gc_rwsem[READ])) { + up_read(&fi->i_gc_rwsem[WRITE]); + ret = -EAGAIN; + goto out; + } + } else { + ret = f2fs_convert_inline_inode(inode); + if (ret) + goto out; + + down_read(&fi->i_gc_rwsem[WRITE]); + if (do_opu) + down_read(&fi->i_gc_rwsem[READ]); + } + if (whint_mode == WHINT_MODE_OFF) + iocb->ki_hint = WRITE_LIFE_NOT_SET; + + /* + * We have to use __iomap_dio_rw() and iomap_dio_complete() instead of + * the higher-level function iomap_dio_rw() in order to ensure that the + * F2FS_DIO_WRITE counter will be decremented correctly in all cases. + */ + inc_page_count(sbi, F2FS_DIO_WRITE); + dio_flags = 0; + if (pos + count > inode->i_size) + dio_flags |= IOMAP_DIO_FORCE_WAIT; + dio = __iomap_dio_rw(iocb, from, &f2fs_iomap_ops, + &f2fs_iomap_dio_write_ops, dio_flags, 0); + if (IS_ERR_OR_NULL(dio)) { + ret = PTR_ERR_OR_ZERO(dio); + if (ret == -ENOTBLK) + ret = 0; + if (ret != -EIOCBQUEUED) + dec_page_count(sbi, F2FS_DIO_WRITE); + } else { + ret = iomap_dio_complete(dio); + } + + if (whint_mode == WHINT_MODE_OFF) + iocb->ki_hint = hint; + if (do_opu) + up_read(&fi->i_gc_rwsem[READ]); + up_read(&fi->i_gc_rwsem[WRITE]); + + if (ret < 0) + goto out; + if (pos + ret > inode->i_size) + f2fs_i_size_write(inode, pos + ret); + if (!do_opu) + set_inode_flag(inode, FI_UPDATE_WRITE); + + if (iov_iter_count(from)) { + ssize_t ret2; + loff_t bufio_start_pos = iocb->ki_pos; + + /* + * The direct write was partial, so we need to fall back to a + * buffered write for the remainder. + */ + + ret2 = f2fs_buffered_write_iter(iocb, from); + if (iov_iter_count(from)) + f2fs_write_failed(inode, iocb->ki_pos); + if (ret2 < 0) + goto out; + + /* + * Ensure that the pagecache pages are written to disk and + * invalidated to preserve the expected O_DIRECT semantics. + */ + if (ret2 > 0) { + loff_t bufio_end_pos = bufio_start_pos + ret2 - 1; + + ret += ret2; + + ret2 = filemap_write_and_wait_range(file->f_mapping, + bufio_start_pos, + bufio_end_pos); + if (ret2 < 0) + goto out; + invalidate_mapping_pages(file->f_mapping, + bufio_start_pos >> PAGE_SHIFT, + bufio_end_pos >> PAGE_SHIFT); + } + } else { + /* iomap_dio_rw() already handled the generic_write_sync(). */ + *may_need_sync = false; + } +out: + trace_f2fs_direct_IO_exit(inode, pos, count, WRITE, ret); + return ret; +} + +static ssize_t f2fs_file_write_iter(struct kiocb *iocb, struct iov_iter *from) +{ + struct inode *inode = file_inode(iocb->ki_filp); + const loff_t orig_pos = iocb->ki_pos; + const size_t orig_count = iov_iter_count(from); + loff_t target_size; + bool dio; + bool may_need_sync = true; + int preallocated; ssize_t ret; if (unlikely(f2fs_cp_error(F2FS_I_SB(inode)))) { @@ -4260,91 +4624,42 @@ static ssize_t f2fs_file_write_iter(struct kiocb *iocb, struct iov_iter *from) inode_lock(inode); } - if (unlikely(IS_IMMUTABLE(inode))) { - ret = -EPERM; - goto unlock; - } - - if (is_inode_flag_set(inode, FI_COMPRESS_RELEASED)) { - ret = -EPERM; - goto unlock; - } - - ret = generic_write_checks(iocb, from); - if (ret > 0) { - bool preallocated = false; - size_t target_size = 0; - int err; - - if (fault_in_iov_iter_readable(from, iov_iter_count(from))) - set_inode_flag(inode, FI_NO_PREALLOC); - - if ((iocb->ki_flags & IOCB_NOWAIT)) { - if (!f2fs_overwrite_io(inode, iocb->ki_pos, - iov_iter_count(from)) || - f2fs_has_inline_data(inode) || - f2fs_force_buffered_io(inode, iocb, from)) { - clear_inode_flag(inode, FI_NO_PREALLOC); - inode_unlock(inode); - ret = -EAGAIN; - goto out; - } - goto write; - } - - if (is_inode_flag_set(inode, FI_NO_PREALLOC)) - goto write; - - if (iocb->ki_flags & IOCB_DIRECT) { - /* - * Convert inline data for Direct I/O before entering - * f2fs_direct_IO(). - */ - err = f2fs_convert_inline_inode(inode); - if (err) - goto out_err; - /* - * If force_buffere_io() is true, we have to allocate - * blocks all the time, since f2fs_direct_IO will fall - * back to buffered IO. - */ - if (!f2fs_force_buffered_io(inode, iocb, from) && - f2fs_lfs_mode(F2FS_I_SB(inode))) - goto write; - } - preallocated = true; - target_size = iocb->ki_pos + iov_iter_count(from); + ret = f2fs_write_checks(iocb, from); + if (ret <= 0) + goto out_unlock; - err = f2fs_preallocate_blocks(iocb, from); - if (err) { -out_err: - clear_inode_flag(inode, FI_NO_PREALLOC); - inode_unlock(inode); - ret = err; - goto out; - } -write: - ret = __generic_file_write_iter(iocb, from); - clear_inode_flag(inode, FI_NO_PREALLOC); + /* Determine whether we will do a direct write or a buffered write. */ + dio = f2fs_should_use_dio(inode, iocb, from); - /* if we couldn't write data, we should deallocate blocks. */ - if (preallocated && i_size_read(inode) < target_size) { - down_write(&F2FS_I(inode)->i_gc_rwsem[WRITE]); - filemap_invalidate_lock(inode->i_mapping); - f2fs_truncate(inode); - filemap_invalidate_unlock(inode->i_mapping); - up_write(&F2FS_I(inode)->i_gc_rwsem[WRITE]); - } + /* Possibly preallocate the blocks for the write. */ + target_size = iocb->ki_pos + iov_iter_count(from); + preallocated = f2fs_preallocate_blocks(iocb, from, dio); + if (preallocated < 0) + ret = preallocated; + else + /* Do the actual write. */ + ret = dio ? + f2fs_dio_write_iter(iocb, from, &may_need_sync): + f2fs_buffered_write_iter(iocb, from); - if (ret > 0) - f2fs_update_iostat(F2FS_I_SB(inode), APP_WRITE_IO, ret); + /* Don't leave any preallocated blocks around past i_size. */ + if (preallocated && i_size_read(inode) < target_size) { + down_write(&F2FS_I(inode)->i_gc_rwsem[WRITE]); + filemap_invalidate_lock(inode->i_mapping); + if (!f2fs_truncate(inode)) + file_dont_truncate(inode); + filemap_invalidate_unlock(inode->i_mapping); + up_write(&F2FS_I(inode)->i_gc_rwsem[WRITE]); + } else { + file_dont_truncate(inode); } -unlock: + + clear_inode_flag(inode, FI_PREALLOCATED_ALL); +out_unlock: inode_unlock(inode); out: - trace_f2fs_file_write_iter(inode, iocb->ki_pos, - iov_iter_count(from), ret); - if (ret > 0) + trace_f2fs_file_write_iter(inode, orig_pos, orig_count, ret); + if (ret > 0 && may_need_sync) ret = generic_write_sync(iocb, ret); return ret; } @@ -4352,12 +4667,12 @@ out: static int f2fs_file_fadvise(struct file *filp, loff_t offset, loff_t len, int advice) { - struct inode *inode; struct address_space *mapping; struct backing_dev_info *bdi; + struct inode *inode = file_inode(filp); + int err; if (advice == POSIX_FADV_SEQUENTIAL) { - inode = file_inode(filp); if (S_ISFIFO(inode->i_mode)) return -ESPIPE; @@ -4374,7 +4689,13 @@ static int f2fs_file_fadvise(struct file *filp, loff_t offset, loff_t len, return 0; } - return generic_fadvise(filp, offset, len, advice); + err = generic_fadvise(filp, offset, len, advice); + if (!err && advice == POSIX_FADV_DONTNEED && + test_opt(F2FS_I_SB(inode), COMPRESS_CACHE) && + f2fs_compressed_file(inode)) + f2fs_invalidate_compress_pages(F2FS_I_SB(inode), inode->i_ino); + + return err; } #ifdef CONFIG_COMPAT diff --git a/fs/f2fs/gc.c b/fs/f2fs/gc.c index a946ce0ead34..ee308a8de432 100644 --- a/fs/f2fs/gc.c +++ b/fs/f2fs/gc.c @@ -7,7 +7,6 @@ */ #include <linux/fs.h> #include <linux/module.h> -#include <linux/backing-dev.h> #include <linux/init.h> #include <linux/f2fs_fs.h> #include <linux/kthread.h> @@ -15,6 +14,7 @@ #include <linux/freezer.h> #include <linux/sched/signal.h> #include <linux/random.h> +#include <linux/sched/mm.h> #include "f2fs.h" #include "node.h" @@ -92,6 +92,18 @@ static int gc_thread_func(void *data) * So, I'd like to wait some time to collect dirty segments. */ if (sbi->gc_mode == GC_URGENT_HIGH) { + spin_lock(&sbi->gc_urgent_high_lock); + if (sbi->gc_urgent_high_limited) { + if (!sbi->gc_urgent_high_remaining) { + sbi->gc_urgent_high_limited = false; + spin_unlock(&sbi->gc_urgent_high_lock); + sbi->gc_mode = GC_NORMAL; + continue; + } + sbi->gc_urgent_high_remaining--; + } + spin_unlock(&sbi->gc_urgent_high_lock); + wait_ms = gc_th->urgent_sleep_time; down_write(&sbi->gc_lock); goto do_gc; @@ -947,7 +959,7 @@ next_step: continue; } - if (f2fs_get_node_info(sbi, nid, &ni)) { + if (f2fs_get_node_info(sbi, nid, &ni, false)) { f2fs_put_page(node_page, 1); continue; } @@ -1015,7 +1027,7 @@ static bool is_alive(struct f2fs_sb_info *sbi, struct f2fs_summary *sum, if (IS_ERR(node_page)) return false; - if (f2fs_get_node_info(sbi, nid, dni)) { + if (f2fs_get_node_info(sbi, nid, dni, false)) { f2fs_put_page(node_page, 1); return false; } @@ -1026,6 +1038,9 @@ static bool is_alive(struct f2fs_sb_info *sbi, struct f2fs_summary *sum, set_sbi_flag(sbi, SBI_NEED_FSCK); } + if (f2fs_check_nid_range(sbi, dni->ino)) + return false; + *nofs = ofs_of_node(node_page); source_blkaddr = data_blkaddr(NULL, node_page, ofs_in_node); f2fs_put_page(node_page, 1); @@ -1039,7 +1054,7 @@ static bool is_alive(struct f2fs_sb_info *sbi, struct f2fs_summary *sum, if (!test_and_set_bit(segno, SIT_I(sbi)->invalid_segmap)) { f2fs_err(sbi, "mismatched blkaddr %u (source_blkaddr %u) in seg %u", blkaddr, source_blkaddr, segno); - f2fs_bug_on(sbi, 1); + set_sbi_flag(sbi, SBI_NEED_FSCK); } } #endif @@ -1206,7 +1221,7 @@ static int move_data_block(struct inode *inode, block_t bidx, f2fs_wait_on_block_writeback(inode, dn.data_blkaddr); - err = f2fs_get_node_info(fio.sbi, dn.nid, &ni); + err = f2fs_get_node_info(fio.sbi, dn.nid, &ni, false); if (err) goto put_out; @@ -1375,8 +1390,7 @@ retry: if (err) { clear_page_private_gcing(page); if (err == -ENOMEM) { - congestion_wait(BLK_RW_ASYNC, - DEFAULT_IO_TIMEOUT); + memalloc_retry_wait(GFP_NOFS); goto retry; } if (is_dirty) @@ -1457,7 +1471,8 @@ next_step: if (phase == 3) { inode = f2fs_iget(sb, dni.ino); - if (IS_ERR(inode) || is_bad_inode(inode)) + if (IS_ERR(inode) || is_bad_inode(inode) || + special_file(inode->i_mode)) continue; if (!down_write_trylock( diff --git a/fs/f2fs/inline.c b/fs/f2fs/inline.c index ea08f0dfa1bd..4b5cefa3f90c 100644 --- a/fs/f2fs/inline.c +++ b/fs/f2fs/inline.c @@ -131,7 +131,7 @@ int f2fs_convert_inline_page(struct dnode_of_data *dn, struct page *page) if (err) return err; - err = f2fs_get_node_info(fio.sbi, dn->nid, &ni); + err = f2fs_get_node_info(fio.sbi, dn->nid, &ni, false); if (err) { f2fs_truncate_data_blocks_range(dn, 1); f2fs_put_dnode(dn); @@ -786,7 +786,7 @@ int f2fs_inline_data_fiemap(struct inode *inode, ilen = start + len; ilen -= start; - err = f2fs_get_node_info(F2FS_I_SB(inode), inode->i_ino, &ni); + err = f2fs_get_node_info(F2FS_I_SB(inode), inode->i_ino, &ni, false); if (err) goto out; diff --git a/fs/f2fs/inode.c b/fs/f2fs/inode.c index 0f8b2df3e1e0..0ec8e32a00b4 100644 --- a/fs/f2fs/inode.c +++ b/fs/f2fs/inode.c @@ -8,8 +8,8 @@ #include <linux/fs.h> #include <linux/f2fs_fs.h> #include <linux/buffer_head.h> -#include <linux/backing-dev.h> #include <linux/writeback.h> +#include <linux/sched/mm.h> #include "f2fs.h" #include "node.h" @@ -516,6 +516,11 @@ make_now: } else if (ino == F2FS_COMPRESS_INO(sbi)) { #ifdef CONFIG_F2FS_FS_COMPRESSION inode->i_mapping->a_ops = &f2fs_compress_aops; + /* + * generic_error_remove_page only truncates pages of regular + * inode + */ + inode->i_mode |= S_IFREG; #endif mapping_set_gfp_mask(inode->i_mapping, GFP_NOFS | __GFP_HIGHMEM | __GFP_MOVABLE); @@ -544,6 +549,14 @@ make_now: goto bad_inode; } f2fs_set_inode_flags(inode); + + if (file_should_truncate(inode)) { + ret = f2fs_truncate(inode); + if (ret) + goto bad_inode; + file_dont_truncate(inode); + } + unlock_new_inode(inode); trace_f2fs_iget(inode); return inode; @@ -562,7 +575,7 @@ retry: inode = f2fs_iget(sb, ino); if (IS_ERR(inode)) { if (PTR_ERR(inode) == -ENOMEM) { - congestion_wait(BLK_RW_ASYNC, DEFAULT_IO_TIMEOUT); + memalloc_retry_wait(GFP_NOFS); goto retry; } } @@ -738,7 +751,8 @@ void f2fs_evict_inode(struct inode *inode) trace_f2fs_evict_inode(inode); truncate_inode_pages_final(&inode->i_data); - if (test_opt(sbi, COMPRESS_CACHE) && f2fs_compressed_file(inode)) + if ((inode->i_nlink || is_bad_inode(inode)) && + test_opt(sbi, COMPRESS_CACHE) && f2fs_compressed_file(inode)) f2fs_invalidate_compress_pages(sbi, inode->i_ino); if (inode->i_ino == F2FS_NODE_INO(sbi) || @@ -868,7 +882,7 @@ void f2fs_handle_failed_inode(struct inode *inode) * so we can prevent losing this orphan when encoutering checkpoint * and following suddenly power-off. */ - err = f2fs_get_node_info(sbi, inode->i_ino, &ni); + err = f2fs_get_node_info(sbi, inode->i_ino, &ni, false); if (err) { set_sbi_flag(sbi, SBI_NEED_FSCK); f2fs_warn(sbi, "May loss orphan inode, run fsck to fix."); diff --git a/fs/f2fs/iostat.c b/fs/f2fs/iostat.c index cdcf54ae0db8..be599f31d3c4 100644 --- a/fs/f2fs/iostat.c +++ b/fs/f2fs/iostat.c @@ -92,7 +92,7 @@ static inline void __record_iostat_latency(struct f2fs_sb_info *sbi) struct f2fs_iostat_latency iostat_lat[MAX_IO_TYPE][NR_PAGE_TYPE]; struct iostat_lat_info *io_lat = sbi->iostat_io_lat; - spin_lock_irq(&sbi->iostat_lat_lock); + spin_lock_bh(&sbi->iostat_lat_lock); for (idx = 0; idx < MAX_IO_TYPE; idx++) { for (io = 0; io < NR_PAGE_TYPE; io++) { cnt = io_lat->bio_cnt[idx][io]; @@ -106,7 +106,7 @@ static inline void __record_iostat_latency(struct f2fs_sb_info *sbi) io_lat->bio_cnt[idx][io] = 0; } } - spin_unlock_irq(&sbi->iostat_lat_lock); + spin_unlock_bh(&sbi->iostat_lat_lock); trace_f2fs_iostat_latency(sbi, iostat_lat); } @@ -120,9 +120,9 @@ static inline void f2fs_record_iostat(struct f2fs_sb_info *sbi) return; /* Need double check under the lock */ - spin_lock(&sbi->iostat_lock); + spin_lock_bh(&sbi->iostat_lock); if (time_is_after_jiffies(sbi->iostat_next_period)) { - spin_unlock(&sbi->iostat_lock); + spin_unlock_bh(&sbi->iostat_lock); return; } sbi->iostat_next_period = jiffies + @@ -133,7 +133,7 @@ static inline void f2fs_record_iostat(struct f2fs_sb_info *sbi) sbi->prev_rw_iostat[i]; sbi->prev_rw_iostat[i] = sbi->rw_iostat[i]; } - spin_unlock(&sbi->iostat_lock); + spin_unlock_bh(&sbi->iostat_lock); trace_f2fs_iostat(sbi, iostat_diff); @@ -145,16 +145,16 @@ void f2fs_reset_iostat(struct f2fs_sb_info *sbi) struct iostat_lat_info *io_lat = sbi->iostat_io_lat; int i; - spin_lock(&sbi->iostat_lock); + spin_lock_bh(&sbi->iostat_lock); for (i = 0; i < NR_IO_TYPE; i++) { sbi->rw_iostat[i] = 0; sbi->prev_rw_iostat[i] = 0; } - spin_unlock(&sbi->iostat_lock); + spin_unlock_bh(&sbi->iostat_lock); - spin_lock_irq(&sbi->iostat_lat_lock); + spin_lock_bh(&sbi->iostat_lat_lock); memset(io_lat, 0, sizeof(struct iostat_lat_info)); - spin_unlock_irq(&sbi->iostat_lat_lock); + spin_unlock_bh(&sbi->iostat_lat_lock); } void f2fs_update_iostat(struct f2fs_sb_info *sbi, @@ -163,19 +163,16 @@ void f2fs_update_iostat(struct f2fs_sb_info *sbi, if (!sbi->iostat_enable) return; - spin_lock(&sbi->iostat_lock); + spin_lock_bh(&sbi->iostat_lock); sbi->rw_iostat[type] += io_bytes; - if (type == APP_WRITE_IO || type == APP_DIRECT_IO) - sbi->rw_iostat[APP_BUFFERED_IO] = - sbi->rw_iostat[APP_WRITE_IO] - - sbi->rw_iostat[APP_DIRECT_IO]; + if (type == APP_BUFFERED_IO || type == APP_DIRECT_IO) + sbi->rw_iostat[APP_WRITE_IO] += io_bytes; - if (type == APP_READ_IO || type == APP_DIRECT_READ_IO) - sbi->rw_iostat[APP_BUFFERED_READ_IO] = - sbi->rw_iostat[APP_READ_IO] - - sbi->rw_iostat[APP_DIRECT_READ_IO]; - spin_unlock(&sbi->iostat_lock); + if (type == APP_BUFFERED_READ_IO || type == APP_DIRECT_READ_IO) + sbi->rw_iostat[APP_READ_IO] += io_bytes; + + spin_unlock_bh(&sbi->iostat_lock); f2fs_record_iostat(sbi); } @@ -185,7 +182,6 @@ static inline void __update_iostat_latency(struct bio_iostat_ctx *iostat_ctx, { unsigned long ts_diff; unsigned int iotype = iostat_ctx->type; - unsigned long flags; struct f2fs_sb_info *sbi = iostat_ctx->sbi; struct iostat_lat_info *io_lat = sbi->iostat_io_lat; int idx; @@ -206,12 +202,12 @@ static inline void __update_iostat_latency(struct bio_iostat_ctx *iostat_ctx, idx = WRITE_ASYNC_IO; } - spin_lock_irqsave(&sbi->iostat_lat_lock, flags); + spin_lock_bh(&sbi->iostat_lat_lock); io_lat->sum_lat[idx][iotype] += ts_diff; io_lat->bio_cnt[idx][iotype]++; if (ts_diff > io_lat->peak_lat[idx][iotype]) io_lat->peak_lat[idx][iotype] = ts_diff; - spin_unlock_irqrestore(&sbi->iostat_lat_lock, flags); + spin_unlock_bh(&sbi->iostat_lat_lock); } void iostat_update_and_unbind_ctx(struct bio *bio, int rw) diff --git a/fs/f2fs/node.c b/fs/f2fs/node.c index 556fcd8457f3..50b2874e758c 100644 --- a/fs/f2fs/node.c +++ b/fs/f2fs/node.c @@ -8,7 +8,7 @@ #include <linux/fs.h> #include <linux/f2fs_fs.h> #include <linux/mpage.h> -#include <linux/backing-dev.h> +#include <linux/sched/mm.h> #include <linux/blkdev.h> #include <linux/pagevec.h> #include <linux/swap.h> @@ -430,6 +430,10 @@ static void cache_nat_entry(struct f2fs_sb_info *sbi, nid_t nid, struct f2fs_nm_info *nm_i = NM_I(sbi); struct nat_entry *new, *e; + /* Let's mitigate lock contention of nat_tree_lock during checkpoint */ + if (rwsem_is_locked(&sbi->cp_global_sem)) + return; + new = __alloc_nat_entry(sbi, nid, false); if (!new) return; @@ -539,7 +543,7 @@ int f2fs_try_to_free_nats(struct f2fs_sb_info *sbi, int nr_shrink) } int f2fs_get_node_info(struct f2fs_sb_info *sbi, nid_t nid, - struct node_info *ni) + struct node_info *ni, bool checkpoint_context) { struct f2fs_nm_info *nm_i = NM_I(sbi); struct curseg_info *curseg = CURSEG_I(sbi, CURSEG_HOT_DATA); @@ -572,9 +576,10 @@ retry: * nat_tree_lock. Therefore, we should retry, if we failed to grab here * while not bothering checkpoint. */ - if (!rwsem_is_locked(&sbi->cp_global_sem)) { + if (!rwsem_is_locked(&sbi->cp_global_sem) || checkpoint_context) { down_read(&curseg->journal_rwsem); - } else if (!down_read_trylock(&curseg->journal_rwsem)) { + } else if (rwsem_is_contended(&nm_i->nat_tree_lock) || + !down_read_trylock(&curseg->journal_rwsem)) { up_read(&nm_i->nat_tree_lock); goto retry; } @@ -887,7 +892,7 @@ static int truncate_node(struct dnode_of_data *dn) int err; pgoff_t index; - err = f2fs_get_node_info(sbi, dn->nid, &ni); + err = f2fs_get_node_info(sbi, dn->nid, &ni, false); if (err) return err; @@ -1286,7 +1291,7 @@ struct page *f2fs_new_node_page(struct dnode_of_data *dn, unsigned int ofs) goto fail; #ifdef CONFIG_F2FS_CHECK_FS - err = f2fs_get_node_info(sbi, dn->nid, &new_ni); + err = f2fs_get_node_info(sbi, dn->nid, &new_ni, false); if (err) { dec_valid_node_count(sbi, dn->inode, !ofs); goto fail; @@ -1348,7 +1353,7 @@ static int read_node_page(struct page *page, int op_flags) return LOCKED_PAGE; } - err = f2fs_get_node_info(sbi, page->index, &ni); + err = f2fs_get_node_info(sbi, page->index, &ni, false); if (err) return err; @@ -1600,7 +1605,7 @@ static int __write_node_page(struct page *page, bool atomic, bool *submitted, nid = nid_of_node(page); f2fs_bug_on(sbi, page->index != nid); - if (f2fs_get_node_info(sbi, nid, &ni)) + if (f2fs_get_node_info(sbi, nid, &ni, !do_balance)) goto redirty_out; if (wbc->for_reclaim) { @@ -2701,7 +2706,7 @@ int f2fs_recover_xattr_data(struct inode *inode, struct page *page) goto recover_xnid; /* 1: invalidate the previous xattr nid */ - err = f2fs_get_node_info(sbi, prev_xnid, &ni); + err = f2fs_get_node_info(sbi, prev_xnid, &ni, false); if (err) return err; @@ -2741,7 +2746,7 @@ int f2fs_recover_inode_page(struct f2fs_sb_info *sbi, struct page *page) struct page *ipage; int err; - err = f2fs_get_node_info(sbi, ino, &old_ni); + err = f2fs_get_node_info(sbi, ino, &old_ni, false); if (err) return err; @@ -2750,7 +2755,7 @@ int f2fs_recover_inode_page(struct f2fs_sb_info *sbi, struct page *page) retry: ipage = f2fs_grab_cache_page(NODE_MAPPING(sbi), ino, false); if (!ipage) { - congestion_wait(BLK_RW_ASYNC, DEFAULT_IO_TIMEOUT); + memalloc_retry_wait(GFP_NOFS); goto retry; } diff --git a/fs/f2fs/recovery.c b/fs/f2fs/recovery.c index 6a1b4668d933..9683c80ff8c2 100644 --- a/fs/f2fs/recovery.c +++ b/fs/f2fs/recovery.c @@ -8,6 +8,7 @@ #include <asm/unaligned.h> #include <linux/fs.h> #include <linux/f2fs_fs.h> +#include <linux/sched/mm.h> #include "f2fs.h" #include "node.h" #include "segment.h" @@ -587,7 +588,7 @@ retry_dn: err = f2fs_get_dnode_of_data(&dn, start, ALLOC_NODE); if (err) { if (err == -ENOMEM) { - congestion_wait(BLK_RW_ASYNC, DEFAULT_IO_TIMEOUT); + memalloc_retry_wait(GFP_NOFS); goto retry_dn; } goto out; @@ -595,7 +596,7 @@ retry_dn: f2fs_wait_on_page_writeback(dn.node_page, NODE, true, true); - err = f2fs_get_node_info(sbi, dn.nid, &ni); + err = f2fs_get_node_info(sbi, dn.nid, &ni, false); if (err) goto err; @@ -670,8 +671,7 @@ retry_prev: err = check_index_in_prev_nodes(sbi, dest, &dn); if (err) { if (err == -ENOMEM) { - congestion_wait(BLK_RW_ASYNC, - DEFAULT_IO_TIMEOUT); + memalloc_retry_wait(GFP_NOFS); goto retry_prev; } goto err; diff --git a/fs/f2fs/segment.c b/fs/f2fs/segment.c index df9ed75f0b7a..1dabc8244083 100644 --- a/fs/f2fs/segment.c +++ b/fs/f2fs/segment.c @@ -9,6 +9,7 @@ #include <linux/f2fs_fs.h> #include <linux/bio.h> #include <linux/blkdev.h> +#include <linux/sched/mm.h> #include <linux/prefetch.h> #include <linux/kthread.h> #include <linux/swap.h> @@ -245,16 +246,14 @@ retry: LOOKUP_NODE); if (err) { if (err == -ENOMEM) { - congestion_wait(BLK_RW_ASYNC, - DEFAULT_IO_TIMEOUT); - cond_resched(); + memalloc_retry_wait(GFP_NOFS); goto retry; } err = -EAGAIN; goto next; } - err = f2fs_get_node_info(sbi, dn.nid, &ni); + err = f2fs_get_node_info(sbi, dn.nid, &ni, false); if (err) { f2fs_put_dnode(&dn); return err; @@ -424,9 +423,7 @@ retry: err = f2fs_do_write_data_page(&fio); if (err) { if (err == -ENOMEM) { - congestion_wait(BLK_RW_ASYNC, - DEFAULT_IO_TIMEOUT); - cond_resched(); + memalloc_retry_wait(GFP_NOFS); goto retry; } unlock_page(page); @@ -2558,8 +2555,8 @@ find_other_zone: secno = find_next_zero_bit(free_i->free_secmap, MAIN_SECS(sbi), hint); if (secno >= MAIN_SECS(sbi)) { if (dir == ALLOC_RIGHT) { - secno = find_next_zero_bit(free_i->free_secmap, - MAIN_SECS(sbi), 0); + secno = find_first_zero_bit(free_i->free_secmap, + MAIN_SECS(sbi)); f2fs_bug_on(sbi, secno >= MAIN_SECS(sbi)); } else { go_left = 1; @@ -2574,8 +2571,8 @@ find_other_zone: left_start--; continue; } - left_start = find_next_zero_bit(free_i->free_secmap, - MAIN_SECS(sbi), 0); + left_start = find_first_zero_bit(free_i->free_secmap, + MAIN_SECS(sbi)); f2fs_bug_on(sbi, left_start >= MAIN_SECS(sbi)); break; } diff --git a/fs/f2fs/segment.h b/fs/f2fs/segment.h index 46fde9f3f28e..0291cd55cf09 100644 --- a/fs/f2fs/segment.h +++ b/fs/f2fs/segment.h @@ -538,7 +538,8 @@ static inline unsigned int free_segments(struct f2fs_sb_info *sbi) static inline unsigned int reserved_segments(struct f2fs_sb_info *sbi) { - return SM_I(sbi)->reserved_segments; + return SM_I(sbi)->reserved_segments + + SM_I(sbi)->additional_reserved_segments; } static inline unsigned int free_sections(struct f2fs_sb_info *sbi) diff --git a/fs/f2fs/super.c b/fs/f2fs/super.c index 040b6d02e1d8..76e6a3df9aba 100644 --- a/fs/f2fs/super.c +++ b/fs/f2fs/super.c @@ -8,9 +8,9 @@ #include <linux/module.h> #include <linux/init.h> #include <linux/fs.h> +#include <linux/sched/mm.h> #include <linux/statfs.h> #include <linux/buffer_head.h> -#include <linux/backing-dev.h> #include <linux/kthread.h> #include <linux/parser.h> #include <linux/mount.h> @@ -59,6 +59,7 @@ const char *f2fs_fault_name[FAULT_MAX] = { [FAULT_WRITE_IO] = "write IO error", [FAULT_SLAB_ALLOC] = "slab alloc", [FAULT_DQUOT_INIT] = "dquot initialize", + [FAULT_LOCK_OP] = "lock_op", }; void f2fs_build_fault_attr(struct f2fs_sb_info *sbi, unsigned int rate, @@ -260,29 +261,22 @@ void f2fs_printk(struct f2fs_sb_info *sbi, const char *fmt, ...) static const struct f2fs_sb_encodings { __u16 magic; char *name; - char *version; + unsigned int version; } f2fs_sb_encoding_map[] = { - {F2FS_ENC_UTF8_12_1, "utf8", "12.1.0"}, + {F2FS_ENC_UTF8_12_1, "utf8", UNICODE_AGE(12, 1, 0)}, }; -static int f2fs_sb_read_encoding(const struct f2fs_super_block *sb, - const struct f2fs_sb_encodings **encoding, - __u16 *flags) +static const struct f2fs_sb_encodings * +f2fs_sb_read_encoding(const struct f2fs_super_block *sb) { __u16 magic = le16_to_cpu(sb->s_encoding); int i; for (i = 0; i < ARRAY_SIZE(f2fs_sb_encoding_map); i++) if (magic == f2fs_sb_encoding_map[i].magic) - break; - - if (i >= ARRAY_SIZE(f2fs_sb_encoding_map)) - return -EINVAL; - - *encoding = &f2fs_sb_encoding_map[i]; - *flags = le16_to_cpu(sb->s_encoding_flags); + return &f2fs_sb_encoding_map[i]; - return 0; + return NULL; } struct kmem_cache *f2fs_cf_name_slab; @@ -328,6 +322,46 @@ static inline void limit_reserve_root(struct f2fs_sb_info *sbi) F2FS_OPTION(sbi).s_resgid)); } +static inline int adjust_reserved_segment(struct f2fs_sb_info *sbi) +{ + unsigned int sec_blks = sbi->blocks_per_seg * sbi->segs_per_sec; + unsigned int avg_vblocks; + unsigned int wanted_reserved_segments; + block_t avail_user_block_count; + + if (!F2FS_IO_ALIGNED(sbi)) + return 0; + + /* average valid block count in section in worst case */ + avg_vblocks = sec_blks / F2FS_IO_SIZE(sbi); + + /* + * we need enough free space when migrating one section in worst case + */ + wanted_reserved_segments = (F2FS_IO_SIZE(sbi) / avg_vblocks) * + reserved_segments(sbi); + wanted_reserved_segments -= reserved_segments(sbi); + + avail_user_block_count = sbi->user_block_count - + sbi->current_reserved_blocks - + F2FS_OPTION(sbi).root_reserved_blocks; + + if (wanted_reserved_segments * sbi->blocks_per_seg > + avail_user_block_count) { + f2fs_err(sbi, "IO align feature can't grab additional reserved segment: %u, available segments: %u", + wanted_reserved_segments, + avail_user_block_count >> sbi->log_blocks_per_seg); + return -ENOSPC; + } + + SM_I(sbi)->additional_reserved_segments = wanted_reserved_segments; + + f2fs_info(sbi, "IO align feature needs additional reserved segment: %u", + wanted_reserved_segments); + + return 0; +} + static inline void adjust_unusable_cap_perc(struct f2fs_sb_info *sbi) { if (!F2FS_OPTION(sbi).unusable_cap_perc) @@ -2415,8 +2449,7 @@ repeat: page = read_cache_page_gfp(mapping, blkidx, GFP_NOFS); if (IS_ERR(page)) { if (PTR_ERR(page) == -ENOMEM) { - congestion_wait(BLK_RW_ASYNC, - DEFAULT_IO_TIMEOUT); + memalloc_retry_wait(GFP_NOFS); goto repeat; } set_sbi_flag(F2FS_SB(sb), SBI_QUOTA_NEED_REPAIR); @@ -3548,6 +3581,7 @@ static void init_sb_info(struct f2fs_sb_info *sbi) sbi->seq_file_ra_mul = MIN_RA_MUL; sbi->max_fragment_chunk = DEF_FRAGMENT_SIZE; sbi->max_fragment_hole = DEF_FRAGMENT_SIZE; + spin_lock_init(&sbi->gc_urgent_high_lock); sbi->dir_level = DEF_DIR_LEVEL; sbi->interval_time[CP_TIME] = DEF_CP_INTERVAL; @@ -3875,25 +3909,32 @@ static int f2fs_setup_casefold(struct f2fs_sb_info *sbi) struct unicode_map *encoding; __u16 encoding_flags; - if (f2fs_sb_read_encoding(sbi->raw_super, &encoding_info, - &encoding_flags)) { + encoding_info = f2fs_sb_read_encoding(sbi->raw_super); + if (!encoding_info) { f2fs_err(sbi, "Encoding requested by superblock is unknown"); return -EINVAL; } + encoding_flags = le16_to_cpu(sbi->raw_super->s_encoding_flags); encoding = utf8_load(encoding_info->version); if (IS_ERR(encoding)) { f2fs_err(sbi, - "can't mount with superblock charset: %s-%s " + "can't mount with superblock charset: %s-%u.%u.%u " "not supported by the kernel. flags: 0x%x.", - encoding_info->name, encoding_info->version, + encoding_info->name, + unicode_major(encoding_info->version), + unicode_minor(encoding_info->version), + unicode_rev(encoding_info->version), encoding_flags); return PTR_ERR(encoding); } f2fs_info(sbi, "Using encoding defined by superblock: " - "%s-%s with flags 0x%hx", encoding_info->name, - encoding_info->version?:"\b", encoding_flags); + "%s-%u.%u.%u with flags 0x%hx", encoding_info->name, + unicode_major(encoding_info->version), + unicode_minor(encoding_info->version), + unicode_rev(encoding_info->version), + encoding_flags); sbi->sb->s_encoding = encoding; sbi->sb->s_encoding_flags = encoding_flags; @@ -4180,6 +4221,10 @@ try_onemore: goto free_nm; } + err = adjust_reserved_segment(sbi); + if (err) + goto free_nm; + /* For write statistics */ sbi->sectors_written_start = f2fs_get_sectors_written(sbi); diff --git a/fs/f2fs/sysfs.c b/fs/f2fs/sysfs.c index 7d289249cd7e..df406c16b2eb 100644 --- a/fs/f2fs/sysfs.c +++ b/fs/f2fs/sysfs.c @@ -118,6 +118,15 @@ static ssize_t sb_status_show(struct f2fs_attr *a, return sprintf(buf, "%lx\n", sbi->s_flag); } +static ssize_t pending_discard_show(struct f2fs_attr *a, + struct f2fs_sb_info *sbi, char *buf) +{ + if (!SM_I(sbi)->dcc_info) + return -EINVAL; + return sprintf(buf, "%llu\n", (unsigned long long)atomic_read( + &SM_I(sbi)->dcc_info->discard_cmd_cnt)); +} + static ssize_t features_show(struct f2fs_attr *a, struct f2fs_sb_info *sbi, char *buf) { @@ -196,8 +205,7 @@ static ssize_t encoding_show(struct f2fs_attr *a, struct super_block *sb = sbi->sb; if (f2fs_sb_has_casefold(sbi)) - return sysfs_emit(buf, "%s (%d.%d.%d)\n", - sb->s_encoding->charset, + return sysfs_emit(buf, "UTF-8 (%d.%d.%d)\n", (sb->s_encoding->version >> 16) & 0xff, (sb->s_encoding->version >> 8) & 0xff, sb->s_encoding->version & 0xff); @@ -415,7 +423,9 @@ out: if (a->struct_type == RESERVED_BLOCKS) { spin_lock(&sbi->stat_lock); if (t > (unsigned long)(sbi->user_block_count - - F2FS_OPTION(sbi).root_reserved_blocks)) { + F2FS_OPTION(sbi).root_reserved_blocks - + sbi->blocks_per_seg * + SM_I(sbi)->additional_reserved_segments)) { spin_unlock(&sbi->stat_lock); return -EINVAL; } @@ -478,6 +488,15 @@ out: return count; } + if (!strcmp(a->attr.name, "gc_urgent_high_remaining")) { + spin_lock(&sbi->gc_urgent_high_lock); + sbi->gc_urgent_high_limited = t != 0; + sbi->gc_urgent_high_remaining = t; + spin_unlock(&sbi->gc_urgent_high_lock); + + return count; + } + #ifdef CONFIG_F2FS_IOSTAT if (!strcmp(a->attr.name, "iostat_enable")) { sbi->iostat_enable = !!t; @@ -733,6 +752,7 @@ F2FS_RW_ATTR(FAULT_INFO_TYPE, f2fs_fault_info, inject_type, inject_type); #endif F2FS_RW_ATTR(F2FS_SBI, f2fs_sb_info, data_io_flag, data_io_flag); F2FS_RW_ATTR(F2FS_SBI, f2fs_sb_info, node_io_flag, node_io_flag); +F2FS_RW_ATTR(F2FS_SBI, f2fs_sb_info, gc_urgent_high_remaining, gc_urgent_high_remaining); F2FS_RW_ATTR(CPRC_INFO, ckpt_req_control, ckpt_thread_ioprio, ckpt_thread_ioprio); F2FS_GENERAL_RO_ATTR(dirty_segments); F2FS_GENERAL_RO_ATTR(free_segments); @@ -744,6 +764,7 @@ F2FS_GENERAL_RO_ATTR(unusable); F2FS_GENERAL_RO_ATTR(encoding); F2FS_GENERAL_RO_ATTR(mounted_time_sec); F2FS_GENERAL_RO_ATTR(main_blkaddr); +F2FS_GENERAL_RO_ATTR(pending_discard); #ifdef CONFIG_F2FS_STAT_FS F2FS_STAT_ATTR(STAT_INFO, f2fs_stat_info, cp_foreground_calls, cp_count); F2FS_STAT_ATTR(STAT_INFO, f2fs_stat_info, cp_background_calls, bg_cp_count); @@ -812,6 +833,7 @@ static struct attribute *f2fs_attrs[] = { ATTR_LIST(main_blkaddr), ATTR_LIST(max_small_discards), ATTR_LIST(discard_granularity), + ATTR_LIST(pending_discard), ATTR_LIST(batched_trim_sections), ATTR_LIST(ipu_policy), ATTR_LIST(min_ipu_util), @@ -844,6 +866,7 @@ static struct attribute *f2fs_attrs[] = { #endif ATTR_LIST(data_io_flag), ATTR_LIST(node_io_flag), + ATTR_LIST(gc_urgent_high_remaining), ATTR_LIST(ckpt_thread_ioprio), ATTR_LIST(dirty_segments), ATTR_LIST(free_segments), diff --git a/fs/f2fs/xattr.c b/fs/f2fs/xattr.c index e348f33bcb2b..8e5cd9c916ff 100644 --- a/fs/f2fs/xattr.c +++ b/fs/f2fs/xattr.c @@ -226,15 +226,18 @@ static inline const struct xattr_handler *f2fs_xattr_handler(int index) } static struct f2fs_xattr_entry *__find_xattr(void *base_addr, - void *last_base_addr, int index, - size_t len, const char *name) + void *last_base_addr, void **last_addr, + int index, size_t len, const char *name) { struct f2fs_xattr_entry *entry; list_for_each_xattr(entry, base_addr) { if ((void *)(entry) + sizeof(__u32) > last_base_addr || - (void *)XATTR_NEXT_ENTRY(entry) > last_base_addr) + (void *)XATTR_NEXT_ENTRY(entry) > last_base_addr) { + if (last_addr) + *last_addr = entry; return NULL; + } if (entry->e_name_index != index) continue; @@ -254,19 +257,9 @@ static struct f2fs_xattr_entry *__find_inline_xattr(struct inode *inode, unsigned int inline_size = inline_xattr_size(inode); void *max_addr = base_addr + inline_size; - list_for_each_xattr(entry, base_addr) { - if ((void *)entry + sizeof(__u32) > max_addr || - (void *)XATTR_NEXT_ENTRY(entry) > max_addr) { - *last_addr = entry; - return NULL; - } - if (entry->e_name_index != index) - continue; - if (entry->e_name_len != len) - continue; - if (!memcmp(entry->e_name, name, len)) - break; - } + entry = __find_xattr(base_addr, max_addr, last_addr, index, len, name); + if (!entry) + return NULL; /* inline xattr header or entry across max inline xattr size */ if (IS_XATTR_LAST_ENTRY(entry) && @@ -368,7 +361,7 @@ static int lookup_all_xattrs(struct inode *inode, struct page *ipage, else cur_addr = txattr_addr; - *xe = __find_xattr(cur_addr, last_txattr_addr, index, len, name); + *xe = __find_xattr(cur_addr, last_txattr_addr, NULL, index, len, name); if (!*xe) { f2fs_err(F2FS_I_SB(inode), "inode (%lu) has corrupted xattr", inode->i_ino); @@ -659,7 +652,7 @@ static int __f2fs_setxattr(struct inode *inode, int index, last_base_addr = (void *)base_addr + XATTR_SIZE(inode); /* find entry with wanted name. */ - here = __find_xattr(base_addr, last_base_addr, index, len, name); + here = __find_xattr(base_addr, last_base_addr, NULL, index, len, name); if (!here) { f2fs_err(F2FS_I_SB(inode), "inode (%lu) has corrupted xattr", inode->i_ino); @@ -684,8 +677,17 @@ static int __f2fs_setxattr(struct inode *inode, int index, } last = here; - while (!IS_XATTR_LAST_ENTRY(last)) + while (!IS_XATTR_LAST_ENTRY(last)) { + if ((void *)(last) + sizeof(__u32) > last_base_addr || + (void *)XATTR_NEXT_ENTRY(last) > last_base_addr) { + f2fs_err(F2FS_I_SB(inode), "inode (%lu) has invalid last xattr entry, entry_size: %zu", + inode->i_ino, ENTRY_SIZE(last)); + set_sbi_flag(F2FS_I_SB(inode), SBI_NEED_FSCK); + error = -EFSCORRUPTED; + goto exit; + } last = XATTR_NEXT_ENTRY(last); + } newsize = XATTR_ALIGN(sizeof(struct f2fs_xattr_entry) + len + size); diff --git a/fs/fat/file.c b/fs/fat/file.c index 13855ba49cd9..a5a309fcc7fa 100644 --- a/fs/fat/file.c +++ b/fs/fat/file.c @@ -175,9 +175,10 @@ long fat_generic_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) static int fat_file_release(struct inode *inode, struct file *filp) { if ((filp->f_mode & FMODE_WRITE) && - MSDOS_SB(inode->i_sb)->options.flush) { + MSDOS_SB(inode->i_sb)->options.flush) { fat_flush_inodes(inode->i_sb, inode, NULL); - congestion_wait(BLK_RW_ASYNC, HZ/10); + set_current_state(TASK_UNINTERRUPTIBLE); + io_schedule_timeout(HZ/10); } return 0; } diff --git a/fs/file_table.c b/fs/file_table.c index 45437f8e1003..57edef16dce4 100644 --- a/fs/file_table.c +++ b/fs/file_table.c @@ -33,7 +33,7 @@ #include "internal.h" /* sysctl tunables... */ -struct files_stat_struct files_stat = { +static struct files_stat_struct files_stat = { .max_files = NR_FILE }; @@ -75,22 +75,53 @@ unsigned long get_max_files(void) } EXPORT_SYMBOL_GPL(get_max_files); +#if defined(CONFIG_SYSCTL) && defined(CONFIG_PROC_FS) + /* * Handle nr_files sysctl */ -#if defined(CONFIG_SYSCTL) && defined(CONFIG_PROC_FS) -int proc_nr_files(struct ctl_table *table, int write, - void *buffer, size_t *lenp, loff_t *ppos) +static int proc_nr_files(struct ctl_table *table, int write, void *buffer, + size_t *lenp, loff_t *ppos) { files_stat.nr_files = get_nr_files(); return proc_doulongvec_minmax(table, write, buffer, lenp, ppos); } -#else -int proc_nr_files(struct ctl_table *table, int write, - void *buffer, size_t *lenp, loff_t *ppos) + +static struct ctl_table fs_stat_sysctls[] = { + { + .procname = "file-nr", + .data = &files_stat, + .maxlen = sizeof(files_stat), + .mode = 0444, + .proc_handler = proc_nr_files, + }, + { + .procname = "file-max", + .data = &files_stat.max_files, + .maxlen = sizeof(files_stat.max_files), + .mode = 0644, + .proc_handler = proc_doulongvec_minmax, + .extra1 = SYSCTL_LONG_ZERO, + .extra2 = SYSCTL_LONG_MAX, + }, + { + .procname = "nr_open", + .data = &sysctl_nr_open, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = &sysctl_nr_open_min, + .extra2 = &sysctl_nr_open_max, + }, + { } +}; + +static int __init init_fs_stat_sysctls(void) { - return -ENOSYS; + register_sysctl_init("fs", fs_stat_sysctls); + return 0; } +fs_initcall(init_fs_stat_sysctls); #endif static struct file *__alloc_file(int flags, const struct cred *cred) diff --git a/fs/fs_context.c b/fs/fs_context.c index b7e43a780a62..24ce12f0db32 100644 --- a/fs/fs_context.c +++ b/fs/fs_context.c @@ -548,7 +548,7 @@ static int legacy_parse_param(struct fs_context *fc, struct fs_parameter *param) param->key); } - if (len > PAGE_SIZE - 2 - size) + if (size + len + 2 > PAGE_SIZE) return invalf(fc, "VFS: Legacy: Cumulative options too large"); if (strchr(param->key, ',') || (param->type == fs_value_is_string && diff --git a/fs/fscache/volume.c b/fs/fscache/volume.c index a57c6cbee858..f2aa7dbad766 100644 --- a/fs/fscache/volume.c +++ b/fs/fscache/volume.c @@ -142,12 +142,12 @@ static void fscache_wait_on_volume_collision(struct fscache_volume *candidate, unsigned int collidee_debug_id) { wait_var_event_timeout(&candidate->flags, - fscache_is_acquire_pending(candidate), 20 * HZ); + !fscache_is_acquire_pending(candidate), 20 * HZ); if (!fscache_is_acquire_pending(candidate)) { pr_notice("Potential volume collision new=%08x old=%08x", candidate->debug_id, collidee_debug_id); fscache_stat(&fscache_n_volumes_collision); - wait_var_event(&candidate->flags, fscache_is_acquire_pending(candidate)); + wait_var_event(&candidate->flags, !fscache_is_acquire_pending(candidate)); } } diff --git a/fs/fuse/virtio_fs.c b/fs/fuse/virtio_fs.c index 74e627109b79..9d737904d07c 100644 --- a/fs/fuse/virtio_fs.c +++ b/fs/fuse/virtio_fs.c @@ -891,7 +891,7 @@ static int virtio_fs_probe(struct virtio_device *vdev) return 0; out_vqs: - vdev->config->reset(vdev); + virtio_reset_device(vdev); virtio_fs_cleanup_vqs(vdev, fs); kfree(fs->vqs); @@ -923,7 +923,7 @@ static void virtio_fs_remove(struct virtio_device *vdev) list_del_init(&fs->list); virtio_fs_stop_all_queues(fs); virtio_fs_drain_all_queues_locked(fs); - vdev->config->reset(vdev); + virtio_reset_device(vdev); virtio_fs_cleanup_vqs(vdev, fs); vdev->priv = NULL; diff --git a/fs/hfsplus/hfsplus_raw.h b/fs/hfsplus/hfsplus_raw.h index 456e87aec7fd..68b4240c6191 100644 --- a/fs/hfsplus/hfsplus_raw.h +++ b/fs/hfsplus/hfsplus_raw.h @@ -260,8 +260,10 @@ struct hfsplus_cat_folder { __be32 access_date; __be32 backup_date; struct hfsplus_perm permissions; - struct DInfo user_info; - struct DXInfo finder_info; + struct_group_attr(info, __packed, + struct DInfo user_info; + struct DXInfo finder_info; + ); __be32 text_encoding; __be32 subfolders; /* Subfolder count in HFSX. Reserved in HFS+. */ } __packed; @@ -294,8 +296,10 @@ struct hfsplus_cat_file { __be32 access_date; __be32 backup_date; struct hfsplus_perm permissions; - struct FInfo user_info; - struct FXInfo finder_info; + struct_group_attr(info, __packed, + struct FInfo user_info; + struct FXInfo finder_info; + ); __be32 text_encoding; u32 reserved2; diff --git a/fs/hfsplus/xattr.c b/fs/hfsplus/xattr.c index e2855ceefd39..49891b12c415 100644 --- a/fs/hfsplus/xattr.c +++ b/fs/hfsplus/xattr.c @@ -296,7 +296,7 @@ int __hfsplus_setxattr(struct inode *inode, const char *name, sizeof(hfsplus_cat_entry)); if (be16_to_cpu(entry.type) == HFSPLUS_FOLDER) { if (size == folder_finderinfo_len) { - memcpy(&entry.folder.user_info, value, + memcpy(&entry.folder.info, value, folder_finderinfo_len); hfs_bnode_write(cat_fd.bnode, &entry, cat_fd.entryoffset, @@ -309,7 +309,7 @@ int __hfsplus_setxattr(struct inode *inode, const char *name, } } else if (be16_to_cpu(entry.type) == HFSPLUS_FILE) { if (size == file_finderinfo_len) { - memcpy(&entry.file.user_info, value, + memcpy(&entry.file.info, value, file_finderinfo_len); hfs_bnode_write(cat_fd.bnode, &entry, cat_fd.entryoffset, diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c index 49d2e686be74..a7c6c7498be0 100644 --- a/fs/hugetlbfs/inode.c +++ b/fs/hugetlbfs/inode.c @@ -409,10 +409,11 @@ hugetlb_vmdelete_list(struct rb_root_cached *root, pgoff_t start, pgoff_t end) struct vm_area_struct *vma; /* - * end == 0 indicates that the entire range after - * start should be unmapped. + * end == 0 indicates that the entire range after start should be + * unmapped. Note, end is exclusive, whereas the interval tree takes + * an inclusive "last". */ - vma_interval_tree_foreach(vma, root, start, end ? end : ULONG_MAX) { + vma_interval_tree_foreach(vma, root, start, end ? end - 1 : ULONG_MAX) { unsigned long v_offset; unsigned long v_end; diff --git a/fs/inode.c b/fs/inode.c index 6b80a51129d5..63324df6fa27 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -67,11 +67,6 @@ const struct address_space_operations empty_aops = { }; EXPORT_SYMBOL(empty_aops); -/* - * Statistics gathering.. - */ -struct inodes_stat_t inodes_stat; - static DEFINE_PER_CPU(unsigned long, nr_inodes); static DEFINE_PER_CPU(unsigned long, nr_unused); @@ -106,13 +101,43 @@ long get_nr_dirty_inodes(void) * Handle nr_inode sysctl */ #ifdef CONFIG_SYSCTL -int proc_nr_inodes(struct ctl_table *table, int write, - void *buffer, size_t *lenp, loff_t *ppos) +/* + * Statistics gathering.. + */ +static struct inodes_stat_t inodes_stat; + +static int proc_nr_inodes(struct ctl_table *table, int write, void *buffer, + size_t *lenp, loff_t *ppos) { inodes_stat.nr_inodes = get_nr_inodes(); inodes_stat.nr_unused = get_nr_inodes_unused(); return proc_doulongvec_minmax(table, write, buffer, lenp, ppos); } + +static struct ctl_table inodes_sysctls[] = { + { + .procname = "inode-nr", + .data = &inodes_stat, + .maxlen = 2*sizeof(long), + .mode = 0444, + .proc_handler = proc_nr_inodes, + }, + { + .procname = "inode-state", + .data = &inodes_stat, + .maxlen = 7*sizeof(long), + .mode = 0444, + .proc_handler = proc_nr_inodes, + }, + { } +}; + +static int __init init_fs_inode_sysctls(void) +{ + register_sysctl_init("fs", inodes_sysctls); + return 0; +} +early_initcall(init_fs_inode_sysctls); #endif static int no_open(struct inode *inode, struct file *file) @@ -526,6 +551,55 @@ void __remove_inode_hash(struct inode *inode) } EXPORT_SYMBOL(__remove_inode_hash); +void dump_mapping(const struct address_space *mapping) +{ + struct inode *host; + const struct address_space_operations *a_ops; + struct hlist_node *dentry_first; + struct dentry *dentry_ptr; + struct dentry dentry; + unsigned long ino; + + /* + * If mapping is an invalid pointer, we don't want to crash + * accessing it, so probe everything depending on it carefully. + */ + if (get_kernel_nofault(host, &mapping->host) || + get_kernel_nofault(a_ops, &mapping->a_ops)) { + pr_warn("invalid mapping:%px\n", mapping); + return; + } + + if (!host) { + pr_warn("aops:%ps\n", a_ops); + return; + } + + if (get_kernel_nofault(dentry_first, &host->i_dentry.first) || + get_kernel_nofault(ino, &host->i_ino)) { + pr_warn("aops:%ps invalid inode:%px\n", a_ops, host); + return; + } + + if (!dentry_first) { + pr_warn("aops:%ps ino:%lx\n", a_ops, ino); + return; + } + + dentry_ptr = container_of(dentry_first, struct dentry, d_u.d_alias); + if (get_kernel_nofault(dentry, dentry_ptr)) { + pr_warn("aops:%ps ino:%lx invalid dentry:%px\n", + a_ops, ino, dentry_ptr); + return; + } + + /* + * if dentry is corrupted, the %pd handler may still crash, + * but it's unlikely that we reach here with a corrupt mapping + */ + pr_warn("aops:%ps ino:%lx dentry name:\"%pd\"\n", a_ops, ino, &dentry); +} + void clear_inode(struct inode *inode) { /* diff --git a/fs/io-wq.c b/fs/io-wq.c index 5c4f582d6549..bb7f161bb19c 100644 --- a/fs/io-wq.c +++ b/fs/io-wq.c @@ -48,7 +48,8 @@ struct io_worker { struct io_wqe *wqe; struct io_wq_work *cur_work; - spinlock_t lock; + struct io_wq_work *next_work; + raw_spinlock_t lock; struct completion ref_done; @@ -405,8 +406,7 @@ static void io_wqe_dec_running(struct io_worker *worker) * Worker will start processing some work. Move it to the busy list, if * it's currently on the freelist */ -static void __io_worker_busy(struct io_wqe *wqe, struct io_worker *worker, - struct io_wq_work *work) +static void __io_worker_busy(struct io_wqe *wqe, struct io_worker *worker) __must_hold(wqe->lock) { if (worker->flags & IO_WORKER_F_FREE) { @@ -529,9 +529,10 @@ static void io_assign_current_work(struct io_worker *worker, cond_resched(); } - spin_lock(&worker->lock); + raw_spin_lock(&worker->lock); worker->cur_work = work; - spin_unlock(&worker->lock); + worker->next_work = NULL; + raw_spin_unlock(&worker->lock); } static void io_wqe_enqueue(struct io_wqe *wqe, struct io_wq_work *work); @@ -546,7 +547,7 @@ static void io_worker_handle_work(struct io_worker *worker) do { struct io_wq_work *work; -get_next: + /* * If we got some work, mark us as busy. If we didn't, but * the list isn't empty, it means we stalled on hashed work. @@ -555,9 +556,20 @@ get_next: * clear the stalled flag. */ work = io_get_next_work(acct, worker); - if (work) - __io_worker_busy(wqe, worker, work); - + if (work) { + __io_worker_busy(wqe, worker); + + /* + * Make sure cancelation can find this, even before + * it becomes the active work. That avoids a window + * where the work has been removed from our general + * work list, but isn't yet discoverable as the + * current work item for this worker. + */ + raw_spin_lock(&worker->lock); + worker->next_work = work; + raw_spin_unlock(&worker->lock); + } raw_spin_unlock(&wqe->lock); if (!work) break; @@ -594,11 +606,6 @@ get_next: spin_unlock_irq(&wq->hash->wait.lock); if (wq_has_sleeper(&wq->hash->wait)) wake_up(&wq->hash->wait); - raw_spin_lock(&wqe->lock); - /* skip unnecessary unlock-lock wqe->lock */ - if (!work) - goto get_next; - raw_spin_unlock(&wqe->lock); } } while (work); @@ -670,7 +677,7 @@ loop: */ void io_wq_worker_running(struct task_struct *tsk) { - struct io_worker *worker = tsk->pf_io_worker; + struct io_worker *worker = tsk->worker_private; if (!worker) return; @@ -688,7 +695,7 @@ void io_wq_worker_running(struct task_struct *tsk) */ void io_wq_worker_sleeping(struct task_struct *tsk) { - struct io_worker *worker = tsk->pf_io_worker; + struct io_worker *worker = tsk->worker_private; if (!worker) return; @@ -707,7 +714,7 @@ void io_wq_worker_sleeping(struct task_struct *tsk) static void io_init_new_worker(struct io_wqe *wqe, struct io_worker *worker, struct task_struct *tsk) { - tsk->pf_io_worker = worker; + tsk->worker_private = worker; worker->task = tsk; set_cpus_allowed_ptr(tsk, wqe->cpu_mask); tsk->flags |= PF_NO_SETAFFINITY; @@ -815,7 +822,7 @@ fail: refcount_set(&worker->ref, 1); worker->wqe = wqe; - spin_lock_init(&worker->lock); + raw_spin_lock_init(&worker->lock); init_completion(&worker->ref_done); if (index == IO_WQ_ACCT_BOUND) @@ -973,6 +980,19 @@ void io_wq_hash_work(struct io_wq_work *work, void *val) work->flags |= (IO_WQ_WORK_HASHED | (bit << IO_WQ_HASH_SHIFT)); } +static bool __io_wq_worker_cancel(struct io_worker *worker, + struct io_cb_cancel_data *match, + struct io_wq_work *work) +{ + if (work && match->fn(work, match->data)) { + work->flags |= IO_WQ_WORK_CANCEL; + set_notify_signal(worker->task); + return true; + } + + return false; +} + static bool io_wq_worker_cancel(struct io_worker *worker, void *data) { struct io_cb_cancel_data *match = data; @@ -981,13 +1001,11 @@ static bool io_wq_worker_cancel(struct io_worker *worker, void *data) * Hold the lock to avoid ->cur_work going out of scope, caller * may dereference the passed in work. */ - spin_lock(&worker->lock); - if (worker->cur_work && - match->fn(worker->cur_work, match->data)) { - set_notify_signal(worker->task); + raw_spin_lock(&worker->lock); + if (__io_wq_worker_cancel(worker, match, worker->cur_work) || + __io_wq_worker_cancel(worker, match, worker->next_work)) match->nr_running++; - } - spin_unlock(&worker->lock); + raw_spin_unlock(&worker->lock); return match->nr_running && !match->cancel_all; } @@ -1039,17 +1057,16 @@ static void io_wqe_cancel_pending_work(struct io_wqe *wqe, { int i; retry: - raw_spin_lock(&wqe->lock); for (i = 0; i < IO_WQ_ACCT_NR; i++) { struct io_wqe_acct *acct = io_get_acct(wqe, i == 0); if (io_acct_cancel_pending_work(wqe, acct, match)) { + raw_spin_lock(&wqe->lock); if (match->cancel_all) goto retry; - return; + break; } } - raw_spin_unlock(&wqe->lock); } static void io_wqe_cancel_running_work(struct io_wqe *wqe, @@ -1074,25 +1091,27 @@ enum io_wq_cancel io_wq_cancel_cb(struct io_wq *wq, work_cancel_fn *cancel, * First check pending list, if we're lucky we can just remove it * from there. CANCEL_OK means that the work is returned as-new, * no completion will be posted for it. - */ - for_each_node(node) { - struct io_wqe *wqe = wq->wqes[node]; - - io_wqe_cancel_pending_work(wqe, &match); - if (match.nr_pending && !match.cancel_all) - return IO_WQ_CANCEL_OK; - } - - /* - * Now check if a free (going busy) or busy worker has the work + * + * Then check if a free (going busy) or busy worker has the work * currently running. If we find it there, we'll return CANCEL_RUNNING * as an indication that we attempt to signal cancellation. The * completion will run normally in this case. + * + * Do both of these while holding the wqe->lock, to ensure that + * we'll find a work item regardless of state. */ for_each_node(node) { struct io_wqe *wqe = wq->wqes[node]; + raw_spin_lock(&wqe->lock); + io_wqe_cancel_pending_work(wqe, &match); + if (match.nr_pending && !match.cancel_all) { + raw_spin_unlock(&wqe->lock); + return IO_WQ_CANCEL_OK; + } + io_wqe_cancel_running_work(wqe, &match); + raw_spin_unlock(&wqe->lock); if (match.nr_running && !match.cancel_all) return IO_WQ_CANCEL_RUNNING; } @@ -1263,7 +1282,9 @@ static void io_wq_destroy(struct io_wq *wq) .fn = io_wq_work_match_all, .cancel_all = true, }; + raw_spin_lock(&wqe->lock); io_wqe_cancel_pending_work(wqe, &match); + raw_spin_unlock(&wqe->lock); free_cpumask_var(wqe->cpu_mask); kfree(wqe); } diff --git a/fs/io-wq.h b/fs/io-wq.h index 3709b7c5ec98..dbecd27656c7 100644 --- a/fs/io-wq.h +++ b/fs/io-wq.h @@ -222,6 +222,6 @@ static inline void io_wq_worker_running(struct task_struct *tsk) static inline bool io_wq_current_is_worker(void) { return in_task() && (current->flags & PF_IO_WORKER) && - current->pf_io_worker; + current->worker_private; } #endif diff --git a/fs/io_uring.c b/fs/io_uring.c index de9c9de90655..e54c4127422e 100644 --- a/fs/io_uring.c +++ b/fs/io_uring.c @@ -1192,12 +1192,6 @@ static inline bool req_ref_put_and_test(struct io_kiocb *req) return atomic_dec_and_test(&req->refs); } -static inline void req_ref_put(struct io_kiocb *req) -{ - WARN_ON_ONCE(!(req->flags & REQ_F_REFCOUNT)); - WARN_ON_ONCE(req_ref_put_and_test(req)); -} - static inline void req_ref_get(struct io_kiocb *req) { WARN_ON_ONCE(!(req->flags & REQ_F_REFCOUNT)); @@ -5468,12 +5462,14 @@ static void io_init_poll_iocb(struct io_poll_iocb *poll, __poll_t events, static inline void io_poll_remove_entry(struct io_poll_iocb *poll) { - struct wait_queue_head *head = poll->head; + struct wait_queue_head *head = smp_load_acquire(&poll->head); - spin_lock_irq(&head->lock); - list_del_init(&poll->wait.entry); - poll->head = NULL; - spin_unlock_irq(&head->lock); + if (head) { + spin_lock_irq(&head->lock); + list_del_init(&poll->wait.entry); + poll->head = NULL; + spin_unlock_irq(&head->lock); + } } static void io_poll_remove_entries(struct io_kiocb *req) @@ -5481,10 +5477,26 @@ static void io_poll_remove_entries(struct io_kiocb *req) struct io_poll_iocb *poll = io_poll_get_single(req); struct io_poll_iocb *poll_double = io_poll_get_double(req); - if (poll->head) - io_poll_remove_entry(poll); - if (poll_double && poll_double->head) + /* + * While we hold the waitqueue lock and the waitqueue is nonempty, + * wake_up_pollfree() will wait for us. However, taking the waitqueue + * lock in the first place can race with the waitqueue being freed. + * + * We solve this as eventpoll does: by taking advantage of the fact that + * all users of wake_up_pollfree() will RCU-delay the actual free. If + * we enter rcu_read_lock() and see that the pointer to the queue is + * non-NULL, we can then lock it without the memory being freed out from + * under us. + * + * Keep holding rcu_read_lock() as long as we hold the queue lock, in + * case the caller deletes the entry from the queue, leaving it empty. + * In that case, only RCU prevents the queue memory from being freed. + */ + rcu_read_lock(); + io_poll_remove_entry(poll); + if (poll_double) io_poll_remove_entry(poll_double); + rcu_read_unlock(); } /* @@ -5624,6 +5636,30 @@ static int io_poll_wake(struct wait_queue_entry *wait, unsigned mode, int sync, wait); __poll_t mask = key_to_poll(key); + if (unlikely(mask & POLLFREE)) { + io_poll_mark_cancelled(req); + /* we have to kick tw in case it's not already */ + io_poll_execute(req, 0); + + /* + * If the waitqueue is being freed early but someone is already + * holds ownership over it, we have to tear down the request as + * best we can. That means immediately removing the request from + * its waitqueue and preventing all further accesses to the + * waitqueue via the request. + */ + list_del_init(&poll->wait.entry); + + /* + * Careful: this *must* be the last step, since as soon + * as req->head is NULL'ed out, the request can be + * completed and freed, since aio_poll_complete_work() + * will no longer need to take the waitqueue lock. + */ + smp_store_release(&poll->head, NULL); + return 1; + } + /* for instances that support it check for an event match first */ if (mask && !(mask & poll->events)) return 0; @@ -6350,16 +6386,21 @@ static int io_try_cancel_userdata(struct io_kiocb *req, u64 sqe_addr) WARN_ON_ONCE(!io_wq_current_is_worker() && req->task != current); ret = io_async_cancel_one(req->task->io_uring, sqe_addr, ctx); - if (ret != -ENOENT) - return ret; + /* + * Fall-through even for -EALREADY, as we may have poll armed + * that need unarming. + */ + if (!ret) + return 0; spin_lock(&ctx->completion_lock); + ret = io_poll_cancel(ctx, sqe_addr, false); + if (ret != -ENOENT) + goto out; + spin_lock_irq(&ctx->timeout_lock); ret = io_timeout_cancel(ctx, sqe_addr); spin_unlock_irq(&ctx->timeout_lock); - if (ret != -ENOENT) - goto out; - ret = io_poll_cancel(ctx, sqe_addr, false); out: spin_unlock(&ctx->completion_lock); return ret; diff --git a/fs/ioctl.c b/fs/ioctl.c index 504e69578112..1ed097e94af2 100644 --- a/fs/ioctl.c +++ b/fs/ioctl.c @@ -430,7 +430,7 @@ static int ioctl_file_dedupe_range(struct file *file, goto out; } - size = offsetof(struct file_dedupe_range __user, info[count]); + size = offsetof(struct file_dedupe_range, info[count]); if (size > PAGE_SIZE) { ret = -ENOMEM; goto out; diff --git a/fs/jbd2/journal.c b/fs/jbd2/journal.c index 0b86a4365b66..f13d548e4a7f 100644 --- a/fs/jbd2/journal.c +++ b/fs/jbd2/journal.c @@ -1212,7 +1212,7 @@ static const struct seq_operations jbd2_seq_info_ops = { static int jbd2_seq_info_open(struct inode *inode, struct file *file) { - journal_t *journal = PDE_DATA(inode); + journal_t *journal = pde_data(inode); struct jbd2_stats_proc_session *s; int rc, size; diff --git a/fs/jffs2/background.c b/fs/jffs2/background.c index 2b4d5013dc5d..6da92ecaf66d 100644 --- a/fs/jffs2/background.c +++ b/fs/jffs2/background.c @@ -161,5 +161,5 @@ static int jffs2_garbage_collect_thread(void *_c) spin_lock(&c->erase_completion_lock); c->gc_task = NULL; spin_unlock(&c->erase_completion_lock); - complete_and_exit(&c->gc_thread_exit, 0); + kthread_complete_and_exit(&c->gc_thread_exit, 0); } diff --git a/fs/ksmbd/asn1.c b/fs/ksmbd/asn1.c index b014f4638610..c03eba090368 100644 --- a/fs/ksmbd/asn1.c +++ b/fs/ksmbd/asn1.c @@ -21,101 +21,11 @@ #include "ksmbd_spnego_negtokeninit.asn1.h" #include "ksmbd_spnego_negtokentarg.asn1.h" -#define SPNEGO_OID_LEN 7 #define NTLMSSP_OID_LEN 10 -#define KRB5_OID_LEN 7 -#define KRB5U2U_OID_LEN 8 -#define MSKRB5_OID_LEN 7 -static unsigned long SPNEGO_OID[7] = { 1, 3, 6, 1, 5, 5, 2 }; -static unsigned long NTLMSSP_OID[10] = { 1, 3, 6, 1, 4, 1, 311, 2, 2, 10 }; -static unsigned long KRB5_OID[7] = { 1, 2, 840, 113554, 1, 2, 2 }; -static unsigned long KRB5U2U_OID[8] = { 1, 2, 840, 113554, 1, 2, 2, 3 }; -static unsigned long MSKRB5_OID[7] = { 1, 2, 840, 48018, 1, 2, 2 }; static char NTLMSSP_OID_STR[NTLMSSP_OID_LEN] = { 0x2b, 0x06, 0x01, 0x04, 0x01, 0x82, 0x37, 0x02, 0x02, 0x0a }; -static bool -asn1_subid_decode(const unsigned char **begin, const unsigned char *end, - unsigned long *subid) -{ - const unsigned char *ptr = *begin; - unsigned char ch; - - *subid = 0; - - do { - if (ptr >= end) - return false; - - ch = *ptr++; - *subid <<= 7; - *subid |= ch & 0x7F; - } while ((ch & 0x80) == 0x80); - - *begin = ptr; - return true; -} - -static bool asn1_oid_decode(const unsigned char *value, size_t vlen, - unsigned long **oid, size_t *oidlen) -{ - const unsigned char *iptr = value, *end = value + vlen; - unsigned long *optr; - unsigned long subid; - - vlen += 1; - if (vlen < 2 || vlen > UINT_MAX / sizeof(unsigned long)) - goto fail_nullify; - - *oid = kmalloc(vlen * sizeof(unsigned long), GFP_KERNEL); - if (!*oid) - return false; - - optr = *oid; - - if (!asn1_subid_decode(&iptr, end, &subid)) - goto fail; - - if (subid < 40) { - optr[0] = 0; - optr[1] = subid; - } else if (subid < 80) { - optr[0] = 1; - optr[1] = subid - 40; - } else { - optr[0] = 2; - optr[1] = subid - 80; - } - - *oidlen = 2; - optr += 2; - - while (iptr < end) { - if (++(*oidlen) > vlen) - goto fail; - - if (!asn1_subid_decode(&iptr, end, optr++)) - goto fail; - } - return true; - -fail: - kfree(*oid); -fail_nullify: - *oid = NULL; - return false; -} - -static bool oid_eq(unsigned long *oid1, unsigned int oid1len, - unsigned long *oid2, unsigned int oid2len) -{ - if (oid1len != oid2len) - return false; - - return memcmp(oid1, oid2, oid1len) == 0; -} - int ksmbd_decode_negTokenInit(unsigned char *security_blob, int length, struct ksmbd_conn *conn) @@ -252,26 +162,18 @@ int build_spnego_ntlmssp_auth_blob(unsigned char **pbuffer, u16 *buflen, int ksmbd_gssapi_this_mech(void *context, size_t hdrlen, unsigned char tag, const void *value, size_t vlen) { - unsigned long *oid; - size_t oidlen; - int err = 0; - - if (!asn1_oid_decode(value, vlen, &oid, &oidlen)) { - err = -EBADMSG; - goto out; - } + enum OID oid; - if (!oid_eq(oid, oidlen, SPNEGO_OID, SPNEGO_OID_LEN)) - err = -EBADMSG; - kfree(oid); -out: - if (err) { + oid = look_up_OID(value, vlen); + if (oid != OID_spnego) { char buf[50]; sprint_oid(value, vlen, buf, sizeof(buf)); ksmbd_debug(AUTH, "Unexpected OID: %s\n", buf); + return -EBADMSG; } - return err; + + return 0; } int ksmbd_neg_token_init_mech_type(void *context, size_t hdrlen, @@ -279,37 +181,31 @@ int ksmbd_neg_token_init_mech_type(void *context, size_t hdrlen, size_t vlen) { struct ksmbd_conn *conn = context; - unsigned long *oid; - size_t oidlen; + enum OID oid; int mech_type; - char buf[50]; - if (!asn1_oid_decode(value, vlen, &oid, &oidlen)) - goto fail; - - if (oid_eq(oid, oidlen, NTLMSSP_OID, NTLMSSP_OID_LEN)) + oid = look_up_OID(value, vlen); + if (oid == OID_ntlmssp) { mech_type = KSMBD_AUTH_NTLMSSP; - else if (oid_eq(oid, oidlen, MSKRB5_OID, MSKRB5_OID_LEN)) + } else if (oid == OID_mskrb5) { mech_type = KSMBD_AUTH_MSKRB5; - else if (oid_eq(oid, oidlen, KRB5_OID, KRB5_OID_LEN)) + } else if (oid == OID_krb5) { mech_type = KSMBD_AUTH_KRB5; - else if (oid_eq(oid, oidlen, KRB5U2U_OID, KRB5U2U_OID_LEN)) + } else if (oid == OID_krb5u2u) { mech_type = KSMBD_AUTH_KRB5U2U; - else - goto fail; + } else { + char buf[50]; + + sprint_oid(value, vlen, buf, sizeof(buf)); + ksmbd_debug(AUTH, "Unexpected OID: %s\n", buf); + return -EBADMSG; + } conn->auth_mechs |= mech_type; if (conn->preferred_auth_mech == 0) conn->preferred_auth_mech = mech_type; - kfree(oid); return 0; - -fail: - kfree(oid); - sprint_oid(value, vlen, buf, sizeof(buf)); - ksmbd_debug(AUTH, "Unexpected OID: %s\n", buf); - return -EBADMSG; } int ksmbd_neg_token_init_mech_token(void *context, size_t hdrlen, diff --git a/fs/ksmbd/auth.c b/fs/ksmbd/auth.c index 3503b1c48cb4..dc3d061edda9 100644 --- a/fs/ksmbd/auth.c +++ b/fs/ksmbd/auth.c @@ -215,7 +215,7 @@ out: * Return: 0 on success, error number on error */ int ksmbd_auth_ntlmv2(struct ksmbd_session *sess, struct ntlmv2_resp *ntlmv2, - int blen, char *domain_name) + int blen, char *domain_name, char *cryptkey) { char ntlmv2_hash[CIFS_ENCPWD_SIZE]; char ntlmv2_rsp[CIFS_HMAC_MD5_HASH_SIZE]; @@ -256,7 +256,7 @@ int ksmbd_auth_ntlmv2(struct ksmbd_session *sess, struct ntlmv2_resp *ntlmv2, goto out; } - memcpy(construct, sess->ntlmssp.cryptkey, CIFS_CRYPTO_KEY_SIZE); + memcpy(construct, cryptkey, CIFS_CRYPTO_KEY_SIZE); memcpy(construct + CIFS_CRYPTO_KEY_SIZE, &ntlmv2->blob_signature, blen); rc = crypto_shash_update(CRYPTO_HMACMD5(ctx), construct, len); @@ -295,7 +295,8 @@ out: * Return: 0 on success, error number on error */ int ksmbd_decode_ntlmssp_auth_blob(struct authenticate_message *authblob, - int blob_len, struct ksmbd_session *sess) + int blob_len, struct ksmbd_conn *conn, + struct ksmbd_session *sess) { char *domain_name; unsigned int nt_off, dn_off; @@ -324,7 +325,7 @@ int ksmbd_decode_ntlmssp_auth_blob(struct authenticate_message *authblob, /* TODO : use domain name that imported from configuration file */ domain_name = smb_strndup_from_utf16((const char *)authblob + dn_off, - dn_len, true, sess->conn->local_nls); + dn_len, true, conn->local_nls); if (IS_ERR(domain_name)) return PTR_ERR(domain_name); @@ -333,7 +334,7 @@ int ksmbd_decode_ntlmssp_auth_blob(struct authenticate_message *authblob, domain_name); ret = ksmbd_auth_ntlmv2(sess, (struct ntlmv2_resp *)((char *)authblob + nt_off), nt_len - CIFS_ENCPWD_SIZE, - domain_name); + domain_name, conn->ntlmssp.cryptkey); kfree(domain_name); return ret; } @@ -347,7 +348,7 @@ int ksmbd_decode_ntlmssp_auth_blob(struct authenticate_message *authblob, * */ int ksmbd_decode_ntlmssp_neg_blob(struct negotiate_message *negblob, - int blob_len, struct ksmbd_session *sess) + int blob_len, struct ksmbd_conn *conn) { if (blob_len < sizeof(struct negotiate_message)) { ksmbd_debug(AUTH, "negotiate blob len %d too small\n", @@ -361,7 +362,7 @@ int ksmbd_decode_ntlmssp_neg_blob(struct negotiate_message *negblob, return -EINVAL; } - sess->ntlmssp.client_flags = le32_to_cpu(negblob->NegotiateFlags); + conn->ntlmssp.client_flags = le32_to_cpu(negblob->NegotiateFlags); return 0; } @@ -375,14 +376,14 @@ int ksmbd_decode_ntlmssp_neg_blob(struct negotiate_message *negblob, */ unsigned int ksmbd_build_ntlmssp_challenge_blob(struct challenge_message *chgblob, - struct ksmbd_session *sess) + struct ksmbd_conn *conn) { struct target_info *tinfo; wchar_t *name; __u8 *target_name; unsigned int flags, blob_off, blob_len, type, target_info_len = 0; int len, uni_len, conv_len; - int cflags = sess->ntlmssp.client_flags; + int cflags = conn->ntlmssp.client_flags; memcpy(chgblob->Signature, NTLMSSP_SIGNATURE, 8); chgblob->MessageType = NtLmChallenge; @@ -403,7 +404,7 @@ ksmbd_build_ntlmssp_challenge_blob(struct challenge_message *chgblob, if (cflags & NTLMSSP_REQUEST_TARGET) flags |= NTLMSSP_REQUEST_TARGET; - if (sess->conn->use_spnego && + if (conn->use_spnego && (cflags & NTLMSSP_NEGOTIATE_EXTENDED_SEC)) flags |= NTLMSSP_NEGOTIATE_EXTENDED_SEC; @@ -414,7 +415,7 @@ ksmbd_build_ntlmssp_challenge_blob(struct challenge_message *chgblob, return -ENOMEM; conv_len = smb_strtoUTF16((__le16 *)name, ksmbd_netbios_name(), len, - sess->conn->local_nls); + conn->local_nls); if (conv_len < 0 || conv_len > len) { kfree(name); return -EINVAL; @@ -430,8 +431,8 @@ ksmbd_build_ntlmssp_challenge_blob(struct challenge_message *chgblob, chgblob->TargetName.BufferOffset = cpu_to_le32(blob_off); /* Initialize random conn challenge */ - get_random_bytes(sess->ntlmssp.cryptkey, sizeof(__u64)); - memcpy(chgblob->Challenge, sess->ntlmssp.cryptkey, + get_random_bytes(conn->ntlmssp.cryptkey, sizeof(__u64)); + memcpy(chgblob->Challenge, conn->ntlmssp.cryptkey, CIFS_CRYPTO_KEY_SIZE); /* Add Target Information to security buffer */ diff --git a/fs/ksmbd/auth.h b/fs/ksmbd/auth.h index 9c2d4badd05d..95629651cf26 100644 --- a/fs/ksmbd/auth.h +++ b/fs/ksmbd/auth.h @@ -38,16 +38,16 @@ struct kvec; int ksmbd_crypt_message(struct ksmbd_conn *conn, struct kvec *iov, unsigned int nvec, int enc); void ksmbd_copy_gss_neg_header(void *buf); -int ksmbd_auth_ntlm(struct ksmbd_session *sess, char *pw_buf); int ksmbd_auth_ntlmv2(struct ksmbd_session *sess, struct ntlmv2_resp *ntlmv2, - int blen, char *domain_name); + int blen, char *domain_name, char *cryptkey); int ksmbd_decode_ntlmssp_auth_blob(struct authenticate_message *authblob, - int blob_len, struct ksmbd_session *sess); + int blob_len, struct ksmbd_conn *conn, + struct ksmbd_session *sess); int ksmbd_decode_ntlmssp_neg_blob(struct negotiate_message *negblob, - int blob_len, struct ksmbd_session *sess); + int blob_len, struct ksmbd_conn *conn); unsigned int ksmbd_build_ntlmssp_challenge_blob(struct challenge_message *chgblob, - struct ksmbd_session *sess); + struct ksmbd_conn *conn); int ksmbd_krb5_authenticate(struct ksmbd_session *sess, char *in_blob, int in_len, char *out_blob, int *out_len); int ksmbd_sign_smb2_pdu(struct ksmbd_conn *conn, char *key, struct kvec *iov, diff --git a/fs/ksmbd/connection.c b/fs/ksmbd/connection.c index 83a94d0bb480..208d2cff7bd3 100644 --- a/fs/ksmbd/connection.c +++ b/fs/ksmbd/connection.c @@ -62,6 +62,7 @@ struct ksmbd_conn *ksmbd_conn_alloc(void) atomic_set(&conn->req_running, 0); atomic_set(&conn->r_count, 0); conn->total_credits = 1; + conn->outstanding_credits = 1; init_waitqueue_head(&conn->req_running_q); INIT_LIST_HEAD(&conn->conns_list); @@ -386,17 +387,24 @@ out: static void stop_sessions(void) { struct ksmbd_conn *conn; + struct ksmbd_transport *t; again: read_lock(&conn_list_lock); list_for_each_entry(conn, &conn_list, conns_list) { struct task_struct *task; - task = conn->transport->handler; + t = conn->transport; + task = t->handler; if (task) ksmbd_debug(CONN, "Stop session handler %s/%d\n", task->comm, task_pid_nr(task)); conn->status = KSMBD_SESS_EXITING; + if (t->ops->shutdown) { + read_unlock(&conn_list_lock); + t->ops->shutdown(t); + read_lock(&conn_list_lock); + } } read_unlock(&conn_list_lock); diff --git a/fs/ksmbd/connection.h b/fs/ksmbd/connection.h index e5403c587a58..7a59aacb5daa 100644 --- a/fs/ksmbd/connection.h +++ b/fs/ksmbd/connection.h @@ -61,8 +61,8 @@ struct ksmbd_conn { atomic_t req_running; /* References which are made for this Server object*/ atomic_t r_count; - unsigned short total_credits; - unsigned short max_credits; + unsigned int total_credits; + unsigned int outstanding_credits; spinlock_t credits_lock; wait_queue_head_t req_running_q; /* Lock to protect requests list*/ @@ -72,12 +72,7 @@ struct ksmbd_conn { int connection_type; struct ksmbd_stats stats; char ClientGUID[SMB2_CLIENT_GUID_SIZE]; - union { - /* pending trans request table */ - struct trans_state *recent_trans; - /* Used by ntlmssp */ - char *ntlmssp_cryptkey; - }; + struct ntlmssp_auth ntlmssp; spinlock_t llist_lock; struct list_head lock_list; @@ -122,6 +117,7 @@ struct ksmbd_conn_ops { struct ksmbd_transport_ops { int (*prepare)(struct ksmbd_transport *t); void (*disconnect)(struct ksmbd_transport *t); + void (*shutdown)(struct ksmbd_transport *t); int (*read)(struct ksmbd_transport *t, char *buf, unsigned int size); int (*writev)(struct ksmbd_transport *t, struct kvec *iovs, int niov, int size, bool need_invalidate_rkey, diff --git a/fs/ksmbd/ksmbd_netlink.h b/fs/ksmbd/ksmbd_netlink.h index c6718a05d347..71bfb7de4472 100644 --- a/fs/ksmbd/ksmbd_netlink.h +++ b/fs/ksmbd/ksmbd_netlink.h @@ -103,6 +103,8 @@ struct ksmbd_startup_request { * we set the SPARSE_FILES bit (0x40). */ __u32 sub_auth[3]; /* Subauth value for Security ID */ + __u32 smb2_max_credits; /* MAX credits */ + __u32 reserved[128]; /* Reserved room */ __u32 ifc_list_sz; /* interfaces list size */ __s8 ____payload[]; }; @@ -113,7 +115,7 @@ struct ksmbd_startup_request { * IPC request to shutdown ksmbd server. */ struct ksmbd_shutdown_request { - __s32 reserved; + __s32 reserved[16]; }; /* @@ -122,6 +124,7 @@ struct ksmbd_shutdown_request { struct ksmbd_login_request { __u32 handle; __s8 account[KSMBD_REQ_MAX_ACCOUNT_NAME_SZ]; /* user account name */ + __u32 reserved[16]; /* Reserved room */ }; /* @@ -135,6 +138,7 @@ struct ksmbd_login_response { __u16 status; __u16 hash_sz; /* hash size */ __s8 hash[KSMBD_REQ_MAX_HASH_SZ]; /* password hash */ + __u32 reserved[16]; /* Reserved room */ }; /* @@ -143,6 +147,7 @@ struct ksmbd_login_response { struct ksmbd_share_config_request { __u32 handle; __s8 share_name[KSMBD_REQ_MAX_SHARE_NAME]; /* share name */ + __u32 reserved[16]; /* Reserved room */ }; /* @@ -157,6 +162,7 @@ struct ksmbd_share_config_response { __u16 force_directory_mode; __u16 force_uid; __u16 force_gid; + __u32 reserved[128]; /* Reserved room */ __u32 veto_list_sz; __s8 ____payload[]; }; @@ -187,6 +193,7 @@ struct ksmbd_tree_connect_request { __s8 account[KSMBD_REQ_MAX_ACCOUNT_NAME_SZ]; __s8 share[KSMBD_REQ_MAX_SHARE_NAME]; __s8 peer_addr[64]; + __u32 reserved[16]; /* Reserved room */ }; /* @@ -196,6 +203,7 @@ struct ksmbd_tree_connect_response { __u32 handle; __u16 status; __u16 connection_flags; + __u32 reserved[16]; /* Reserved room */ }; /* @@ -204,6 +212,7 @@ struct ksmbd_tree_connect_response { struct ksmbd_tree_disconnect_request { __u64 session_id; /* session id */ __u64 connect_id; /* tree connection id */ + __u32 reserved[16]; /* Reserved room */ }; /* @@ -212,6 +221,7 @@ struct ksmbd_tree_disconnect_request { struct ksmbd_logout_request { __s8 account[KSMBD_REQ_MAX_ACCOUNT_NAME_SZ]; /* user account name */ __u32 account_flags; + __u32 reserved[16]; /* Reserved room */ }; /* diff --git a/fs/ksmbd/mgmt/user_config.c b/fs/ksmbd/mgmt/user_config.c index 1019d3677d55..279d00feff21 100644 --- a/fs/ksmbd/mgmt/user_config.c +++ b/fs/ksmbd/mgmt/user_config.c @@ -67,3 +67,13 @@ int ksmbd_anonymous_user(struct ksmbd_user *user) return 1; return 0; } + +bool ksmbd_compare_user(struct ksmbd_user *u1, struct ksmbd_user *u2) +{ + if (strcmp(u1->name, u2->name)) + return false; + if (memcmp(u1->passkey, u2->passkey, u1->passkey_sz)) + return false; + + return true; +} diff --git a/fs/ksmbd/mgmt/user_config.h b/fs/ksmbd/mgmt/user_config.h index aff80b029579..6a44109617f1 100644 --- a/fs/ksmbd/mgmt/user_config.h +++ b/fs/ksmbd/mgmt/user_config.h @@ -64,4 +64,5 @@ struct ksmbd_user *ksmbd_login_user(const char *account); struct ksmbd_user *ksmbd_alloc_user(struct ksmbd_login_response *resp); void ksmbd_free_user(struct ksmbd_user *user); int ksmbd_anonymous_user(struct ksmbd_user *user); +bool ksmbd_compare_user(struct ksmbd_user *u1, struct ksmbd_user *u2); #endif /* __USER_CONFIG_MANAGEMENT_H__ */ diff --git a/fs/ksmbd/mgmt/user_session.h b/fs/ksmbd/mgmt/user_session.h index 82289c3cbd2b..e241f16a3851 100644 --- a/fs/ksmbd/mgmt/user_session.h +++ b/fs/ksmbd/mgmt/user_session.h @@ -45,7 +45,6 @@ struct ksmbd_session { int state; __u8 *Preauth_HashValue; - struct ntlmssp_auth ntlmssp; char sess_key[CIFS_KEY_SIZE]; struct hlist_node hlist; diff --git a/fs/ksmbd/smb2misc.c b/fs/ksmbd/smb2misc.c index 50d0b1022289..4a9460153b59 100644 --- a/fs/ksmbd/smb2misc.c +++ b/fs/ksmbd/smb2misc.c @@ -289,7 +289,7 @@ static int smb2_validate_credit_charge(struct ksmbd_conn *conn, unsigned int req_len = 0, expect_resp_len = 0, calc_credit_num, max_len; unsigned short credit_charge = le16_to_cpu(hdr->CreditCharge); void *__hdr = hdr; - int ret; + int ret = 0; switch (hdr->Command) { case SMB2_QUERY_INFO: @@ -326,21 +326,27 @@ static int smb2_validate_credit_charge(struct ksmbd_conn *conn, ksmbd_debug(SMB, "Insufficient credit charge, given: %d, needed: %d\n", credit_charge, calc_credit_num); return 1; - } else if (credit_charge > conn->max_credits) { + } else if (credit_charge > conn->vals->max_credits) { ksmbd_debug(SMB, "Too large credit charge: %d\n", credit_charge); return 1; } spin_lock(&conn->credits_lock); - if (credit_charge <= conn->total_credits) { - conn->total_credits -= credit_charge; - ret = 0; - } else { + if (credit_charge > conn->total_credits) { ksmbd_debug(SMB, "Insufficient credits granted, given: %u, granted: %u\n", credit_charge, conn->total_credits); ret = 1; } + + if ((u64)conn->outstanding_credits + credit_charge > conn->vals->max_credits) { + ksmbd_debug(SMB, "Limits exceeding the maximum allowable outstanding requests, given : %u, pending : %u\n", + credit_charge, conn->outstanding_credits); + ret = 1; + } else + conn->outstanding_credits += credit_charge; + spin_unlock(&conn->credits_lock); + return ret; } diff --git a/fs/ksmbd/smb2ops.c b/fs/ksmbd/smb2ops.c index 02a44d28bdaf..ab23da2120b9 100644 --- a/fs/ksmbd/smb2ops.c +++ b/fs/ksmbd/smb2ops.c @@ -19,6 +19,7 @@ static struct smb_version_values smb21_server_values = { .max_read_size = SMB21_DEFAULT_IOSIZE, .max_write_size = SMB21_DEFAULT_IOSIZE, .max_trans_size = SMB21_DEFAULT_IOSIZE, + .max_credits = SMB2_MAX_CREDITS, .large_lock_type = 0, .exclusive_lock_type = SMB2_LOCKFLAG_EXCLUSIVE, .shared_lock_type = SMB2_LOCKFLAG_SHARED, @@ -44,6 +45,7 @@ static struct smb_version_values smb30_server_values = { .max_read_size = SMB3_DEFAULT_IOSIZE, .max_write_size = SMB3_DEFAULT_IOSIZE, .max_trans_size = SMB3_DEFAULT_TRANS_SIZE, + .max_credits = SMB2_MAX_CREDITS, .large_lock_type = 0, .exclusive_lock_type = SMB2_LOCKFLAG_EXCLUSIVE, .shared_lock_type = SMB2_LOCKFLAG_SHARED, @@ -70,6 +72,7 @@ static struct smb_version_values smb302_server_values = { .max_read_size = SMB3_DEFAULT_IOSIZE, .max_write_size = SMB3_DEFAULT_IOSIZE, .max_trans_size = SMB3_DEFAULT_TRANS_SIZE, + .max_credits = SMB2_MAX_CREDITS, .large_lock_type = 0, .exclusive_lock_type = SMB2_LOCKFLAG_EXCLUSIVE, .shared_lock_type = SMB2_LOCKFLAG_SHARED, @@ -96,6 +99,7 @@ static struct smb_version_values smb311_server_values = { .max_read_size = SMB3_DEFAULT_IOSIZE, .max_write_size = SMB3_DEFAULT_IOSIZE, .max_trans_size = SMB3_DEFAULT_TRANS_SIZE, + .max_credits = SMB2_MAX_CREDITS, .large_lock_type = 0, .exclusive_lock_type = SMB2_LOCKFLAG_EXCLUSIVE, .shared_lock_type = SMB2_LOCKFLAG_SHARED, @@ -197,7 +201,6 @@ void init_smb2_1_server(struct ksmbd_conn *conn) conn->ops = &smb2_0_server_ops; conn->cmds = smb2_0_server_cmds; conn->max_cmds = ARRAY_SIZE(smb2_0_server_cmds); - conn->max_credits = SMB2_MAX_CREDITS; conn->signing_algorithm = SIGNING_ALG_HMAC_SHA256_LE; if (server_conf.flags & KSMBD_GLOBAL_FLAG_SMB2_LEASES) @@ -215,7 +218,6 @@ void init_smb3_0_server(struct ksmbd_conn *conn) conn->ops = &smb3_0_server_ops; conn->cmds = smb2_0_server_cmds; conn->max_cmds = ARRAY_SIZE(smb2_0_server_cmds); - conn->max_credits = SMB2_MAX_CREDITS; conn->signing_algorithm = SIGNING_ALG_AES_CMAC_LE; if (server_conf.flags & KSMBD_GLOBAL_FLAG_SMB2_LEASES) @@ -240,7 +242,6 @@ void init_smb3_02_server(struct ksmbd_conn *conn) conn->ops = &smb3_0_server_ops; conn->cmds = smb2_0_server_cmds; conn->max_cmds = ARRAY_SIZE(smb2_0_server_cmds); - conn->max_credits = SMB2_MAX_CREDITS; conn->signing_algorithm = SIGNING_ALG_AES_CMAC_LE; if (server_conf.flags & KSMBD_GLOBAL_FLAG_SMB2_LEASES) @@ -265,7 +266,6 @@ int init_smb3_11_server(struct ksmbd_conn *conn) conn->ops = &smb3_11_server_ops; conn->cmds = smb2_0_server_cmds; conn->max_cmds = ARRAY_SIZE(smb2_0_server_cmds); - conn->max_credits = SMB2_MAX_CREDITS; conn->signing_algorithm = SIGNING_ALG_AES_CMAC_LE; if (server_conf.flags & KSMBD_GLOBAL_FLAG_SMB2_LEASES) @@ -304,3 +304,11 @@ void init_smb2_max_trans_size(unsigned int sz) smb302_server_values.max_trans_size = sz; smb311_server_values.max_trans_size = sz; } + +void init_smb2_max_credits(unsigned int sz) +{ + smb21_server_values.max_credits = sz; + smb30_server_values.max_credits = sz; + smb302_server_values.max_credits = sz; + smb311_server_values.max_credits = sz; +} diff --git a/fs/ksmbd/smb2pdu.c b/fs/ksmbd/smb2pdu.c index b8b3a4c28b74..1866c81c5c99 100644 --- a/fs/ksmbd/smb2pdu.c +++ b/fs/ksmbd/smb2pdu.c @@ -299,16 +299,15 @@ int smb2_set_rsp_credits(struct ksmbd_work *work) struct smb2_hdr *req_hdr = ksmbd_req_buf_next(work); struct smb2_hdr *hdr = ksmbd_resp_buf_next(work); struct ksmbd_conn *conn = work->conn; - unsigned short credits_requested; + unsigned short credits_requested, aux_max; unsigned short credit_charge, credits_granted = 0; - unsigned short aux_max, aux_credits; if (work->send_no_response) return 0; hdr->CreditCharge = req_hdr->CreditCharge; - if (conn->total_credits > conn->max_credits) { + if (conn->total_credits > conn->vals->max_credits) { hdr->CreditRequest = 0; pr_err("Total credits overflow: %d\n", conn->total_credits); return -EINVAL; @@ -316,6 +315,14 @@ int smb2_set_rsp_credits(struct ksmbd_work *work) credit_charge = max_t(unsigned short, le16_to_cpu(req_hdr->CreditCharge), 1); + if (credit_charge > conn->total_credits) { + ksmbd_debug(SMB, "Insufficient credits granted, given: %u, granted: %u\n", + credit_charge, conn->total_credits); + return -EINVAL; + } + + conn->total_credits -= credit_charge; + conn->outstanding_credits -= credit_charge; credits_requested = max_t(unsigned short, le16_to_cpu(req_hdr->CreditRequest), 1); @@ -325,16 +332,14 @@ int smb2_set_rsp_credits(struct ksmbd_work *work) * TODO: Need to adjuct CreditRequest value according to * current cpu load */ - aux_credits = credits_requested - 1; if (hdr->Command == SMB2_NEGOTIATE) - aux_max = 0; + aux_max = 1; else - aux_max = conn->max_credits - credit_charge; - aux_credits = min_t(unsigned short, aux_credits, aux_max); - credits_granted = credit_charge + aux_credits; + aux_max = conn->vals->max_credits - credit_charge; + credits_granted = min_t(unsigned short, credits_requested, aux_max); - if (conn->max_credits - conn->total_credits < credits_granted) - credits_granted = conn->max_credits - + if (conn->vals->max_credits - conn->total_credits < credits_granted) + credits_granted = conn->vals->max_credits - conn->total_credits; conn->total_credits += credits_granted; @@ -610,16 +615,14 @@ static void destroy_previous_session(struct ksmbd_user *user, u64 id) /** * smb2_get_name() - get filename string from on the wire smb format - * @share: ksmbd_share_config pointer * @src: source buffer * @maxlen: maxlen of source string - * @nls_table: nls_table pointer + * @local_nls: nls_table pointer * * Return: matching converted filename on success, otherwise error ptr */ static char * -smb2_get_name(struct ksmbd_share_config *share, const char *src, - const int maxlen, struct nls_table *local_nls) +smb2_get_name(const char *src, const int maxlen, struct nls_table *local_nls) { char *name; @@ -1303,7 +1306,7 @@ static int ntlm_negotiate(struct ksmbd_work *work, int sz, rc; ksmbd_debug(SMB, "negotiate phase\n"); - rc = ksmbd_decode_ntlmssp_neg_blob(negblob, negblob_len, work->sess); + rc = ksmbd_decode_ntlmssp_neg_blob(negblob, negblob_len, work->conn); if (rc) return rc; @@ -1313,7 +1316,7 @@ static int ntlm_negotiate(struct ksmbd_work *work, memset(chgblob, 0, sizeof(struct challenge_message)); if (!work->conn->use_spnego) { - sz = ksmbd_build_ntlmssp_challenge_blob(chgblob, work->sess); + sz = ksmbd_build_ntlmssp_challenge_blob(chgblob, work->conn); if (sz < 0) return -ENOMEM; @@ -1329,7 +1332,7 @@ static int ntlm_negotiate(struct ksmbd_work *work, return -ENOMEM; chgblob = (struct challenge_message *)neg_blob; - sz = ksmbd_build_ntlmssp_challenge_blob(chgblob, work->sess); + sz = ksmbd_build_ntlmssp_challenge_blob(chgblob, work->conn); if (sz < 0) { rc = -ENOMEM; goto out; @@ -1450,60 +1453,62 @@ static int ntlm_authenticate(struct ksmbd_work *work) ksmbd_free_user(user); return 0; } - ksmbd_free_user(sess->user); - } - sess->user = user; - if (user_guest(sess->user)) { - if (conn->sign) { - ksmbd_debug(SMB, "Guest login not allowed when signing enabled\n"); + if (!ksmbd_compare_user(sess->user, user)) { + ksmbd_free_user(user); return -EPERM; } + ksmbd_free_user(user); + } else { + sess->user = user; + } + if (user_guest(sess->user)) { rsp->SessionFlags = SMB2_SESSION_FLAG_IS_GUEST_LE; } else { struct authenticate_message *authblob; authblob = user_authblob(conn, req); sz = le16_to_cpu(req->SecurityBufferLength); - rc = ksmbd_decode_ntlmssp_auth_blob(authblob, sz, sess); + rc = ksmbd_decode_ntlmssp_auth_blob(authblob, sz, conn, sess); if (rc) { set_user_flag(sess->user, KSMBD_USER_FLAG_BAD_PASSWORD); ksmbd_debug(SMB, "authentication failed\n"); return -EPERM; } + } - /* - * If session state is SMB2_SESSION_VALID, We can assume - * that it is reauthentication. And the user/password - * has been verified, so return it here. - */ - if (sess->state == SMB2_SESSION_VALID) { - if (conn->binding) - goto binding_session; - return 0; - } + /* + * If session state is SMB2_SESSION_VALID, We can assume + * that it is reauthentication. And the user/password + * has been verified, so return it here. + */ + if (sess->state == SMB2_SESSION_VALID) { + if (conn->binding) + goto binding_session; + return 0; + } - if ((conn->sign || server_conf.enforced_signing) || - (req->SecurityMode & SMB2_NEGOTIATE_SIGNING_REQUIRED)) - sess->sign = true; + if ((rsp->SessionFlags != SMB2_SESSION_FLAG_IS_GUEST_LE && + (conn->sign || server_conf.enforced_signing)) || + (req->SecurityMode & SMB2_NEGOTIATE_SIGNING_REQUIRED)) + sess->sign = true; - if (smb3_encryption_negotiated(conn) && - !(req->Flags & SMB2_SESSION_REQ_FLAG_BINDING)) { - rc = conn->ops->generate_encryptionkey(sess); - if (rc) { - ksmbd_debug(SMB, - "SMB3 encryption key generation failed\n"); - return -EINVAL; - } - sess->enc = true; - rsp->SessionFlags = SMB2_SESSION_FLAG_ENCRYPT_DATA_LE; - /* - * signing is disable if encryption is enable - * on this session - */ - sess->sign = false; + if (smb3_encryption_negotiated(conn) && + !(req->Flags & SMB2_SESSION_REQ_FLAG_BINDING)) { + rc = conn->ops->generate_encryptionkey(sess); + if (rc) { + ksmbd_debug(SMB, + "SMB3 encryption key generation failed\n"); + return -EINVAL; } + sess->enc = true; + rsp->SessionFlags = SMB2_SESSION_FLAG_ENCRYPT_DATA_LE; + /* + * signing is disable if encryption is enable + * on this session + */ + sess->sign = false; } binding_session: @@ -2057,9 +2062,6 @@ int smb2_session_logoff(struct ksmbd_work *work) ksmbd_debug(SMB, "request\n"); - /* Got a valid session, set connection state */ - WARN_ON(sess->conn != conn); - /* setting CifsExiting here may race with start_tcp_sess */ ksmbd_conn_set_need_reconnect(work); ksmbd_close_session_fds(work); @@ -2530,8 +2532,7 @@ int smb2_open(struct ksmbd_work *work) goto err_out1; } - name = smb2_get_name(share, - req->Buffer, + name = smb2_get_name(req->Buffer, le16_to_cpu(req->NameLength), work->conn->local_nls); if (IS_ERR(name)) { @@ -3392,7 +3393,6 @@ static int dentry_name(struct ksmbd_dir_info *d_info, int info_level) * @conn: connection instance * @info_level: smb information level * @d_info: structure included variables for query dir - * @user_ns: user namespace * @ksmbd_kstat: ksmbd wrapper of dirent stat information * * if directory has many entries, find first can't read it fully. @@ -4018,6 +4018,7 @@ err_out2: * buffer_check_err() - helper function to check buffer errors * @reqOutputBufferLength: max buffer length expected in command response * @rsp: query info response buffer contains output buffer length + * @rsp_org: base response buffer pointer in case of chained response * @infoclass_size: query info class response buffer size * * Return: 0 on success, otherwise error @@ -5398,8 +5399,7 @@ static int smb2_rename(struct ksmbd_work *work, goto out; } - new_name = smb2_get_name(share, - file_info->FileName, + new_name = smb2_get_name(file_info->FileName, le32_to_cpu(file_info->FileNameLength), local_nls); if (IS_ERR(new_name)) { @@ -5510,8 +5510,7 @@ static int smb2_create_link(struct ksmbd_work *work, if (!pathname) return -ENOMEM; - link_name = smb2_get_name(share, - file_info->FileName, + link_name = smb2_get_name(file_info->FileName, le32_to_cpu(file_info->FileNameLength), local_nls); if (IS_ERR(link_name) || S_ISDIR(file_inode(filp)->i_mode)) { @@ -5849,7 +5848,7 @@ static int set_file_mode_info(struct ksmbd_file *fp, * smb2_set_info_file() - handler for smb2 set info command * @work: smb work containing set info command buffer * @fp: ksmbd_file pointer - * @info_class: smb2 set info class + * @req: request buffer pointer * @share: ksmbd_share_config pointer * * Return: 0 on success, otherwise error @@ -6121,25 +6120,33 @@ out: return err; } -static ssize_t smb2_read_rdma_channel(struct ksmbd_work *work, - struct smb2_read_req *req, void *data_buf, - size_t length) +static int smb2_set_remote_key_for_rdma(struct ksmbd_work *work, + struct smb2_buffer_desc_v1 *desc, + __le32 Channel, + __le16 ChannelInfoOffset, + __le16 ChannelInfoLength) { - struct smb2_buffer_desc_v1 *desc = - (struct smb2_buffer_desc_v1 *)&req->Buffer[0]; - int err; - if (work->conn->dialect == SMB30_PROT_ID && - req->Channel != SMB2_CHANNEL_RDMA_V1) + Channel != SMB2_CHANNEL_RDMA_V1) return -EINVAL; - if (req->ReadChannelInfoOffset == 0 || - le16_to_cpu(req->ReadChannelInfoLength) < sizeof(*desc)) + if (ChannelInfoOffset == 0 || + le16_to_cpu(ChannelInfoLength) < sizeof(*desc)) return -EINVAL; work->need_invalidate_rkey = - (req->Channel == SMB2_CHANNEL_RDMA_V1_INVALIDATE); + (Channel == SMB2_CHANNEL_RDMA_V1_INVALIDATE); work->remote_key = le32_to_cpu(desc->token); + return 0; +} + +static ssize_t smb2_read_rdma_channel(struct ksmbd_work *work, + struct smb2_read_req *req, void *data_buf, + size_t length) +{ + struct smb2_buffer_desc_v1 *desc = + (struct smb2_buffer_desc_v1 *)&req->Buffer[0]; + int err; err = ksmbd_conn_rdma_write(work->conn, data_buf, length, le32_to_cpu(desc->token), @@ -6162,7 +6169,7 @@ int smb2_read(struct ksmbd_work *work) struct ksmbd_conn *conn = work->conn; struct smb2_read_req *req; struct smb2_read_rsp *rsp; - struct ksmbd_file *fp; + struct ksmbd_file *fp = NULL; loff_t offset; size_t length, mincount; ssize_t nbytes = 0, remain_bytes = 0; @@ -6176,6 +6183,18 @@ int smb2_read(struct ksmbd_work *work) return smb2_read_pipe(work); } + if (req->Channel == SMB2_CHANNEL_RDMA_V1_INVALIDATE || + req->Channel == SMB2_CHANNEL_RDMA_V1) { + err = smb2_set_remote_key_for_rdma(work, + (struct smb2_buffer_desc_v1 *) + &req->Buffer[0], + req->Channel, + req->ReadChannelInfoOffset, + req->ReadChannelInfoLength); + if (err) + goto out; + } + fp = ksmbd_lookup_fd_slow(work, le64_to_cpu(req->VolatileFileId), le64_to_cpu(req->PersistentFileId)); if (!fp) { @@ -6361,21 +6380,6 @@ static ssize_t smb2_write_rdma_channel(struct ksmbd_work *work, desc = (struct smb2_buffer_desc_v1 *)&req->Buffer[0]; - if (work->conn->dialect == SMB30_PROT_ID && - req->Channel != SMB2_CHANNEL_RDMA_V1) - return -EINVAL; - - if (req->Length != 0 || req->DataOffset != 0) - return -EINVAL; - - if (req->WriteChannelInfoOffset == 0 || - le16_to_cpu(req->WriteChannelInfoLength) < sizeof(*desc)) - return -EINVAL; - - work->need_invalidate_rkey = - (req->Channel == SMB2_CHANNEL_RDMA_V1_INVALIDATE); - work->remote_key = le32_to_cpu(desc->token); - data_buf = kvmalloc(length, GFP_KERNEL | __GFP_ZERO); if (!data_buf) return -ENOMEM; @@ -6422,6 +6426,20 @@ int smb2_write(struct ksmbd_work *work) return smb2_write_pipe(work); } + if (req->Channel == SMB2_CHANNEL_RDMA_V1 || + req->Channel == SMB2_CHANNEL_RDMA_V1_INVALIDATE) { + if (req->Length != 0 || req->DataOffset != 0) + return -EINVAL; + err = smb2_set_remote_key_for_rdma(work, + (struct smb2_buffer_desc_v1 *) + &req->Buffer[0], + req->Channel, + req->WriteChannelInfoOffset, + req->WriteChannelInfoLength); + if (err) + goto out; + } + if (!test_tree_conn_flag(work->tcon, KSMBD_TREE_CONN_FLAG_WRITABLE)) { ksmbd_debug(SMB, "User does not have write permission\n"); err = -EACCES; @@ -7243,15 +7261,10 @@ static int fsctl_query_iface_info_ioctl(struct ksmbd_conn *conn, struct sockaddr_storage_rsp *sockaddr_storage; unsigned int flags; unsigned long long speed; - struct sockaddr_in6 *csin6 = (struct sockaddr_in6 *)&conn->peer_addr; rtnl_lock(); for_each_netdev(&init_net, netdev) { - if (out_buf_len < - nbytes + sizeof(struct network_interface_info_ioctl_rsp)) { - rtnl_unlock(); - return -ENOSPC; - } + bool ipv4_set = false; if (netdev->type == ARPHRD_LOOPBACK) continue; @@ -7259,12 +7272,20 @@ static int fsctl_query_iface_info_ioctl(struct ksmbd_conn *conn, flags = dev_get_flags(netdev); if (!(flags & IFF_RUNNING)) continue; +ipv6_retry: + if (out_buf_len < + nbytes + sizeof(struct network_interface_info_ioctl_rsp)) { + rtnl_unlock(); + return -ENOSPC; + } nii_rsp = (struct network_interface_info_ioctl_rsp *) &rsp->Buffer[nbytes]; nii_rsp->IfIndex = cpu_to_le32(netdev->ifindex); nii_rsp->Capability = 0; + if (netdev->real_num_tx_queues > 1) + nii_rsp->Capability |= cpu_to_le32(RSS_CAPABLE); if (ksmbd_rdma_capable_netdev(netdev)) nii_rsp->Capability |= cpu_to_le32(RDMA_CAPABLE); @@ -7289,8 +7310,7 @@ static int fsctl_query_iface_info_ioctl(struct ksmbd_conn *conn, nii_rsp->SockAddr_Storage; memset(sockaddr_storage, 0, 128); - if (conn->peer_addr.ss_family == PF_INET || - ipv6_addr_v4mapped(&csin6->sin6_addr)) { + if (!ipv4_set) { struct in_device *idev; sockaddr_storage->Family = cpu_to_le16(INTERNETWORK); @@ -7301,6 +7321,9 @@ static int fsctl_query_iface_info_ioctl(struct ksmbd_conn *conn, continue; sockaddr_storage->addr4.IPv4address = idev_ipv4_address(idev); + nbytes += sizeof(struct network_interface_info_ioctl_rsp); + ipv4_set = true; + goto ipv6_retry; } else { struct inet6_dev *idev6; struct inet6_ifaddr *ifa; @@ -7322,9 +7345,8 @@ static int fsctl_query_iface_info_ioctl(struct ksmbd_conn *conn, break; } sockaddr_storage->addr6.ScopeId = 0; + nbytes += sizeof(struct network_interface_info_ioctl_rsp); } - - nbytes += sizeof(struct network_interface_info_ioctl_rsp); } rtnl_unlock(); diff --git a/fs/ksmbd/smb2pdu.h b/fs/ksmbd/smb2pdu.h index 4a3e4339d4c4..725b800c29c8 100644 --- a/fs/ksmbd/smb2pdu.h +++ b/fs/ksmbd/smb2pdu.h @@ -980,6 +980,7 @@ int init_smb3_11_server(struct ksmbd_conn *conn); void init_smb2_max_read_size(unsigned int sz); void init_smb2_max_write_size(unsigned int sz); void init_smb2_max_trans_size(unsigned int sz); +void init_smb2_max_credits(unsigned int sz); bool is_smb2_neg_cmd(struct ksmbd_work *work); bool is_smb2_rsp(struct ksmbd_work *work); diff --git a/fs/ksmbd/smb_common.h b/fs/ksmbd/smb_common.h index 50590842b651..e1369b4345a9 100644 --- a/fs/ksmbd/smb_common.h +++ b/fs/ksmbd/smb_common.h @@ -365,6 +365,7 @@ struct smb_version_values { __u32 max_read_size; __u32 max_write_size; __u32 max_trans_size; + __u32 max_credits; __u32 large_lock_type; __u32 exclusive_lock_type; __u32 shared_lock_type; diff --git a/fs/ksmbd/transport_ipc.c b/fs/ksmbd/transport_ipc.c index 1acf1892a466..3ad6881e0f7e 100644 --- a/fs/ksmbd/transport_ipc.c +++ b/fs/ksmbd/transport_ipc.c @@ -301,6 +301,8 @@ static int ipc_server_config_on_startup(struct ksmbd_startup_request *req) init_smb2_max_write_size(req->smb2_max_write); if (req->smb2_max_trans) init_smb2_max_trans_size(req->smb2_max_trans); + if (req->smb2_max_credits) + init_smb2_max_credits(req->smb2_max_credits); ret = ksmbd_set_netbios_name(req->netbios_name); ret |= ksmbd_set_server_string(req->server_string); diff --git a/fs/ksmbd/transport_rdma.c b/fs/ksmbd/transport_rdma.c index 7e57cbb0bb35..3c1ec1ac0b27 100644 --- a/fs/ksmbd/transport_rdma.c +++ b/fs/ksmbd/transport_rdma.c @@ -34,7 +34,8 @@ #include "smbstatus.h" #include "transport_rdma.h" -#define SMB_DIRECT_PORT 5445 +#define SMB_DIRECT_PORT_IWARP 5445 +#define SMB_DIRECT_PORT_INFINIBAND 445 #define SMB_DIRECT_VERSION_LE cpu_to_le16(0x0100) @@ -60,6 +61,10 @@ * as defined in [MS-SMBD] 3.1.1.1 * Those may change after a SMB_DIRECT negotiation */ + +/* Set 445 port to SMB Direct port by default */ +static int smb_direct_port = SMB_DIRECT_PORT_INFINIBAND; + /* The local peer's maximum number of credits to grant to the peer */ static int smb_direct_receive_credit_max = 255; @@ -75,10 +80,18 @@ static int smb_direct_max_fragmented_recv_size = 1024 * 1024; /* The maximum single-message size which can be received */ static int smb_direct_max_receive_size = 8192; -static int smb_direct_max_read_write_size = 1024 * 1024; +static int smb_direct_max_read_write_size = 1048512; static int smb_direct_max_outstanding_rw_ops = 8; +static LIST_HEAD(smb_direct_device_list); +static DEFINE_RWLOCK(smb_direct_device_lock); + +struct smb_direct_device { + struct ib_device *ib_dev; + struct list_head list; +}; + static struct smb_direct_listener { struct rdma_cm_id *cm_id; } smb_direct_listener; @@ -415,6 +428,7 @@ static void free_transport(struct smb_direct_transport *t) if (t->qp) { ib_drain_qp(t->qp); + ib_mr_pool_destroy(t->qp, &t->qp->rdma_mrs); ib_destroy_qp(t->qp); } @@ -555,6 +569,7 @@ static void recv_done(struct ib_cq *cq, struct ib_wc *wc) } t->negotiation_requested = true; t->full_packet_received = true; + enqueue_reassembly(t, recvmsg, 0); wake_up_interruptible(&t->wait_status); break; case SMB_DIRECT_MSG_DATA_TRANSFER: { @@ -1438,6 +1453,15 @@ static void smb_direct_disconnect(struct ksmbd_transport *t) free_transport(st); } +static void smb_direct_shutdown(struct ksmbd_transport *t) +{ + struct smb_direct_transport *st = smb_trans_direct_transfort(t); + + ksmbd_debug(RDMA, "smb-direct shutdown cm_id=%p\n", st->cm_id); + + smb_direct_disconnect_rdma_work(&st->disconnect_work); +} + static int smb_direct_cm_handler(struct rdma_cm_id *cm_id, struct rdma_cm_event *event) { @@ -1581,19 +1605,13 @@ static int smb_direct_accept_client(struct smb_direct_transport *t) pr_err("error at rdma_accept: %d\n", ret); return ret; } - - wait_event_interruptible(t->wait_status, - t->status != SMB_DIRECT_CS_NEW); - if (t->status != SMB_DIRECT_CS_CONNECTED) - return -ENOTCONN; return 0; } -static int smb_direct_negotiate(struct smb_direct_transport *t) +static int smb_direct_prepare_negotiation(struct smb_direct_transport *t) { int ret; struct smb_direct_recvmsg *recvmsg; - struct smb_direct_negotiate_req *req; recvmsg = get_free_recvmsg(t); if (!recvmsg) @@ -1603,44 +1621,20 @@ static int smb_direct_negotiate(struct smb_direct_transport *t) ret = smb_direct_post_recv(t, recvmsg); if (ret) { pr_err("Can't post recv: %d\n", ret); - goto out; + goto out_err; } t->negotiation_requested = false; ret = smb_direct_accept_client(t); if (ret) { pr_err("Can't accept client\n"); - goto out; + goto out_err; } smb_direct_post_recv_credits(&t->post_recv_credits_work.work); - - ksmbd_debug(RDMA, "Waiting for SMB_DIRECT negotiate request\n"); - ret = wait_event_interruptible_timeout(t->wait_status, - t->negotiation_requested || - t->status == SMB_DIRECT_CS_DISCONNECTED, - SMB_DIRECT_NEGOTIATE_TIMEOUT * HZ); - if (ret <= 0 || t->status == SMB_DIRECT_CS_DISCONNECTED) { - ret = ret < 0 ? ret : -ETIMEDOUT; - goto out; - } - - ret = smb_direct_check_recvmsg(recvmsg); - if (ret == -ECONNABORTED) - goto out; - - req = (struct smb_direct_negotiate_req *)recvmsg->packet; - t->max_recv_size = min_t(int, t->max_recv_size, - le32_to_cpu(req->preferred_send_size)); - t->max_send_size = min_t(int, t->max_send_size, - le32_to_cpu(req->max_receive_size)); - t->max_fragmented_send_size = - le32_to_cpu(req->max_fragmented_size); - - ret = smb_direct_send_negotiate_response(t, ret); -out: - if (recvmsg) - put_recvmsg(t, recvmsg); + return 0; +out_err: + put_recvmsg(t, recvmsg); return ret; } @@ -1724,7 +1718,9 @@ static int smb_direct_init_params(struct smb_direct_transport *t, cap->max_send_sge = SMB_DIRECT_MAX_SEND_SGES; cap->max_recv_sge = SMB_DIRECT_MAX_RECV_SGES; cap->max_inline_data = 0; - cap->max_rdma_ctxs = 0; + cap->max_rdma_ctxs = + rdma_rw_mr_factor(device, t->cm_id->port_num, max_pages) * + smb_direct_max_outstanding_rw_ops; return 0; } @@ -1806,6 +1802,7 @@ static int smb_direct_create_qpair(struct smb_direct_transport *t, { int ret; struct ib_qp_init_attr qp_attr; + int pages_per_rw; t->pd = ib_alloc_pd(t->cm_id->device, 0); if (IS_ERR(t->pd)) { @@ -1853,6 +1850,23 @@ static int smb_direct_create_qpair(struct smb_direct_transport *t, t->qp = t->cm_id->qp; t->cm_id->event_handler = smb_direct_cm_handler; + pages_per_rw = DIV_ROUND_UP(t->max_rdma_rw_size, PAGE_SIZE) + 1; + if (pages_per_rw > t->cm_id->device->attrs.max_sgl_rd) { + int pages_per_mr, mr_count; + + pages_per_mr = min_t(int, pages_per_rw, + t->cm_id->device->attrs.max_fast_reg_page_list_len); + mr_count = DIV_ROUND_UP(pages_per_rw, pages_per_mr) * + atomic_read(&t->rw_avail_ops); + ret = ib_mr_pool_init(t->qp, &t->qp->rdma_mrs, mr_count, + IB_MR_TYPE_MEM_REG, pages_per_mr, 0); + if (ret) { + pr_err("failed to init mr pool count %d pages %d\n", + mr_count, pages_per_mr); + goto err; + } + } + return 0; err: if (t->qp) { @@ -1877,6 +1891,49 @@ err: static int smb_direct_prepare(struct ksmbd_transport *t) { struct smb_direct_transport *st = smb_trans_direct_transfort(t); + struct smb_direct_recvmsg *recvmsg; + struct smb_direct_negotiate_req *req; + int ret; + + ksmbd_debug(RDMA, "Waiting for SMB_DIRECT negotiate request\n"); + ret = wait_event_interruptible_timeout(st->wait_status, + st->negotiation_requested || + st->status == SMB_DIRECT_CS_DISCONNECTED, + SMB_DIRECT_NEGOTIATE_TIMEOUT * HZ); + if (ret <= 0 || st->status == SMB_DIRECT_CS_DISCONNECTED) + return ret < 0 ? ret : -ETIMEDOUT; + + recvmsg = get_first_reassembly(st); + if (!recvmsg) + return -ECONNABORTED; + + ret = smb_direct_check_recvmsg(recvmsg); + if (ret == -ECONNABORTED) + goto out; + + req = (struct smb_direct_negotiate_req *)recvmsg->packet; + st->max_recv_size = min_t(int, st->max_recv_size, + le32_to_cpu(req->preferred_send_size)); + st->max_send_size = min_t(int, st->max_send_size, + le32_to_cpu(req->max_receive_size)); + st->max_fragmented_send_size = + le32_to_cpu(req->max_fragmented_size); + st->max_fragmented_recv_size = + (st->recv_credit_max * st->max_recv_size) / 2; + + ret = smb_direct_send_negotiate_response(st, ret); +out: + spin_lock_irq(&st->reassembly_queue_lock); + st->reassembly_queue_length--; + list_del(&recvmsg->list); + spin_unlock_irq(&st->reassembly_queue_lock); + put_recvmsg(st, recvmsg); + + return ret; +} + +static int smb_direct_connect(struct smb_direct_transport *st) +{ int ret; struct ib_qp_cap qp_cap; @@ -1898,13 +1955,11 @@ static int smb_direct_prepare(struct ksmbd_transport *t) return ret; } - ret = smb_direct_negotiate(st); + ret = smb_direct_prepare_negotiation(st); if (ret) { pr_err("Can't negotiate: %d\n", ret); return ret; } - - st->status = SMB_DIRECT_CS_CONNECTED; return 0; } @@ -1920,6 +1975,7 @@ static bool rdma_frwr_is_supported(struct ib_device_attr *attrs) static int smb_direct_handle_connect_request(struct rdma_cm_id *new_cm_id) { struct smb_direct_transport *t; + int ret; if (!rdma_frwr_is_supported(&new_cm_id->device->attrs)) { ksmbd_debug(RDMA, @@ -1932,18 +1988,23 @@ static int smb_direct_handle_connect_request(struct rdma_cm_id *new_cm_id) if (!t) return -ENOMEM; + ret = smb_direct_connect(t); + if (ret) + goto out_err; + KSMBD_TRANS(t)->handler = kthread_run(ksmbd_conn_handler_loop, KSMBD_TRANS(t)->conn, "ksmbd:r%u", - SMB_DIRECT_PORT); + smb_direct_port); if (IS_ERR(KSMBD_TRANS(t)->handler)) { - int ret = PTR_ERR(KSMBD_TRANS(t)->handler); - + ret = PTR_ERR(KSMBD_TRANS(t)->handler); pr_err("Can't start thread\n"); - free_transport(t); - return ret; + goto out_err; } return 0; +out_err: + free_transport(t); + return ret; } static int smb_direct_listen_handler(struct rdma_cm_id *cm_id, @@ -2007,12 +2068,65 @@ err: return ret; } +static int smb_direct_ib_client_add(struct ib_device *ib_dev) +{ + struct smb_direct_device *smb_dev; + + /* Set 5445 port if device type is iWARP(No IB) */ + if (ib_dev->node_type != RDMA_NODE_IB_CA) + smb_direct_port = SMB_DIRECT_PORT_IWARP; + + if (!ib_dev->ops.get_netdev || + !rdma_frwr_is_supported(&ib_dev->attrs)) + return 0; + + smb_dev = kzalloc(sizeof(*smb_dev), GFP_KERNEL); + if (!smb_dev) + return -ENOMEM; + smb_dev->ib_dev = ib_dev; + + write_lock(&smb_direct_device_lock); + list_add(&smb_dev->list, &smb_direct_device_list); + write_unlock(&smb_direct_device_lock); + + ksmbd_debug(RDMA, "ib device added: name %s\n", ib_dev->name); + return 0; +} + +static void smb_direct_ib_client_remove(struct ib_device *ib_dev, + void *client_data) +{ + struct smb_direct_device *smb_dev, *tmp; + + write_lock(&smb_direct_device_lock); + list_for_each_entry_safe(smb_dev, tmp, &smb_direct_device_list, list) { + if (smb_dev->ib_dev == ib_dev) { + list_del(&smb_dev->list); + kfree(smb_dev); + break; + } + } + write_unlock(&smb_direct_device_lock); +} + +static struct ib_client smb_direct_ib_client = { + .name = "ksmbd_smb_direct_ib", + .add = smb_direct_ib_client_add, + .remove = smb_direct_ib_client_remove, +}; + int ksmbd_rdma_init(void) { int ret; smb_direct_listener.cm_id = NULL; + ret = ib_register_client(&smb_direct_ib_client); + if (ret) { + pr_err("failed to ib_register_client\n"); + return ret; + } + /* When a client is running out of send credits, the credits are * granted by the server's sending a packet using this queue. * This avoids the situation that a clients cannot send packets @@ -2023,7 +2137,7 @@ int ksmbd_rdma_init(void) if (!smb_direct_wq) return -ENOMEM; - ret = smb_direct_listen(SMB_DIRECT_PORT); + ret = smb_direct_listen(smb_direct_port); if (ret) { destroy_workqueue(smb_direct_wq); smb_direct_wq = NULL; @@ -2036,36 +2150,67 @@ int ksmbd_rdma_init(void) return 0; } -int ksmbd_rdma_destroy(void) +void ksmbd_rdma_destroy(void) { - if (smb_direct_listener.cm_id) - rdma_destroy_id(smb_direct_listener.cm_id); + if (!smb_direct_listener.cm_id) + return; + + ib_unregister_client(&smb_direct_ib_client); + rdma_destroy_id(smb_direct_listener.cm_id); + smb_direct_listener.cm_id = NULL; if (smb_direct_wq) { destroy_workqueue(smb_direct_wq); smb_direct_wq = NULL; } - return 0; } bool ksmbd_rdma_capable_netdev(struct net_device *netdev) { - struct ib_device *ibdev; + struct smb_direct_device *smb_dev; + int i; bool rdma_capable = false; - ibdev = ib_device_get_by_netdev(netdev, RDMA_DRIVER_UNKNOWN); - if (ibdev) { - if (rdma_frwr_is_supported(&ibdev->attrs)) - rdma_capable = true; - ib_device_put(ibdev); + read_lock(&smb_direct_device_lock); + list_for_each_entry(smb_dev, &smb_direct_device_list, list) { + for (i = 0; i < smb_dev->ib_dev->phys_port_cnt; i++) { + struct net_device *ndev; + + ndev = smb_dev->ib_dev->ops.get_netdev(smb_dev->ib_dev, + i + 1); + if (!ndev) + continue; + + if (ndev == netdev) { + dev_put(ndev); + rdma_capable = true; + goto out; + } + dev_put(ndev); + } + } +out: + read_unlock(&smb_direct_device_lock); + + if (rdma_capable == false) { + struct ib_device *ibdev; + + ibdev = ib_device_get_by_netdev(netdev, RDMA_DRIVER_UNKNOWN); + if (ibdev) { + if (rdma_frwr_is_supported(&ibdev->attrs)) + rdma_capable = true; + ib_device_put(ibdev); + } } + return rdma_capable; } static struct ksmbd_transport_ops ksmbd_smb_direct_transport_ops = { .prepare = smb_direct_prepare, .disconnect = smb_direct_disconnect, + .shutdown = smb_direct_shutdown, .writev = smb_direct_writev, .read = smb_direct_read, .rdma_read = smb_direct_rdma_read, diff --git a/fs/ksmbd/transport_rdma.h b/fs/ksmbd/transport_rdma.h index 0fa8adc0776f..5567d93a6f96 100644 --- a/fs/ksmbd/transport_rdma.h +++ b/fs/ksmbd/transport_rdma.h @@ -7,8 +7,6 @@ #ifndef __KSMBD_TRANSPORT_RDMA_H__ #define __KSMBD_TRANSPORT_RDMA_H__ -#define SMB_DIRECT_PORT 5445 - /* SMB DIRECT negotiation request packet [MS-SMBD] 2.2.1 */ struct smb_direct_negotiate_req { __le16 min_version; @@ -52,7 +50,7 @@ struct smb_direct_data_transfer { #ifdef CONFIG_SMB_SERVER_SMBDIRECT int ksmbd_rdma_init(void); -int ksmbd_rdma_destroy(void); +void ksmbd_rdma_destroy(void); bool ksmbd_rdma_capable_netdev(struct net_device *netdev); #else static inline int ksmbd_rdma_init(void) { return 0; } diff --git a/fs/ksmbd/transport_tcp.c b/fs/ksmbd/transport_tcp.c index c14320e03b69..82a1429bbe12 100644 --- a/fs/ksmbd/transport_tcp.c +++ b/fs/ksmbd/transport_tcp.c @@ -404,7 +404,7 @@ static int create_socket(struct interface *iface) &ksmbd_socket); if (ret) { pr_err("Can't create socket for ipv4: %d\n", ret); - goto out_error; + goto out_clear; } sin.sin_family = PF_INET; @@ -462,6 +462,7 @@ static int create_socket(struct interface *iface) out_error: tcp_destroy_socket(ksmbd_socket); +out_clear: iface->ksmbd_socket = NULL; return ret; } diff --git a/fs/ksmbd/vfs_cache.h b/fs/ksmbd/vfs_cache.h index 448576fbe4b7..36239ce31afd 100644 --- a/fs/ksmbd/vfs_cache.h +++ b/fs/ksmbd/vfs_cache.h @@ -96,16 +96,6 @@ struct ksmbd_file { int durable_timeout; - /* for SMB1 */ - int pid; - - /* conflict lock fail count for SMB1 */ - unsigned int cflock_cnt; - /* last lock failure start offset for SMB1 */ - unsigned long long llock_fstart; - - int dirent_offset; - /* if ls is happening on directory, below is valid*/ struct ksmbd_readdir_data readdir_data; int dot_dotdot[2]; diff --git a/fs/lockd/svc.c b/fs/lockd/svc.c index b220e1b91726..0475c5a5d061 100644 --- a/fs/lockd/svc.c +++ b/fs/lockd/svc.c @@ -54,13 +54,9 @@ EXPORT_SYMBOL_GPL(nlmsvc_ops); static DEFINE_MUTEX(nlmsvc_mutex); static unsigned int nlmsvc_users; -static struct task_struct *nlmsvc_task; -static struct svc_rqst *nlmsvc_rqst; +static struct svc_serv *nlmsvc_serv; unsigned long nlmsvc_timeout; -static atomic_t nlm_ntf_refcnt = ATOMIC_INIT(0); -static DECLARE_WAIT_QUEUE_HEAD(nlm_ntf_wq); - unsigned int lockd_net_id; /* @@ -184,7 +180,12 @@ lockd(void *vrqstp) nlm_shutdown_hosts(); cancel_delayed_work_sync(&ln->grace_period_end); locks_end_grace(&ln->lockd_manager); - return 0; + + dprintk("lockd_down: service stopped\n"); + + svc_exit_thread(rqstp); + + module_put_and_kthread_exit(0); } static int create_lockd_listener(struct svc_serv *serv, const char *name, @@ -290,8 +291,8 @@ static void lockd_down_net(struct svc_serv *serv, struct net *net) __func__, net->ns.inum); } } else { - pr_err("%s: no users! task=%p, net=%x\n", - __func__, nlmsvc_task, net->ns.inum); + pr_err("%s: no users! net=%x\n", + __func__, net->ns.inum); BUG(); } } @@ -302,20 +303,16 @@ static int lockd_inetaddr_event(struct notifier_block *this, struct in_ifaddr *ifa = (struct in_ifaddr *)ptr; struct sockaddr_in sin; - if ((event != NETDEV_DOWN) || - !atomic_inc_not_zero(&nlm_ntf_refcnt)) + if (event != NETDEV_DOWN) goto out; - if (nlmsvc_rqst) { + if (nlmsvc_serv) { dprintk("lockd_inetaddr_event: removed %pI4\n", &ifa->ifa_local); sin.sin_family = AF_INET; sin.sin_addr.s_addr = ifa->ifa_local; - svc_age_temp_xprts_now(nlmsvc_rqst->rq_server, - (struct sockaddr *)&sin); + svc_age_temp_xprts_now(nlmsvc_serv, (struct sockaddr *)&sin); } - atomic_dec(&nlm_ntf_refcnt); - wake_up(&nlm_ntf_wq); out: return NOTIFY_DONE; @@ -332,21 +329,17 @@ static int lockd_inet6addr_event(struct notifier_block *this, struct inet6_ifaddr *ifa = (struct inet6_ifaddr *)ptr; struct sockaddr_in6 sin6; - if ((event != NETDEV_DOWN) || - !atomic_inc_not_zero(&nlm_ntf_refcnt)) + if (event != NETDEV_DOWN) goto out; - if (nlmsvc_rqst) { + if (nlmsvc_serv) { dprintk("lockd_inet6addr_event: removed %pI6\n", &ifa->addr); sin6.sin6_family = AF_INET6; sin6.sin6_addr = ifa->addr; if (ipv6_addr_type(&sin6.sin6_addr) & IPV6_ADDR_LINKLOCAL) sin6.sin6_scope_id = ifa->idev->dev->ifindex; - svc_age_temp_xprts_now(nlmsvc_rqst->rq_server, - (struct sockaddr *)&sin6); + svc_age_temp_xprts_now(nlmsvc_serv, (struct sockaddr *)&sin6); } - atomic_dec(&nlm_ntf_refcnt); - wake_up(&nlm_ntf_wq); out: return NOTIFY_DONE; @@ -357,86 +350,22 @@ static struct notifier_block lockd_inet6addr_notifier = { }; #endif -static void lockd_unregister_notifiers(void) -{ - unregister_inetaddr_notifier(&lockd_inetaddr_notifier); -#if IS_ENABLED(CONFIG_IPV6) - unregister_inet6addr_notifier(&lockd_inet6addr_notifier); -#endif - wait_event(nlm_ntf_wq, atomic_read(&nlm_ntf_refcnt) == 0); -} - -static void lockd_svc_exit_thread(void) -{ - atomic_dec(&nlm_ntf_refcnt); - lockd_unregister_notifiers(); - svc_exit_thread(nlmsvc_rqst); -} - -static int lockd_start_svc(struct svc_serv *serv) -{ - int error; - - if (nlmsvc_rqst) - return 0; - - /* - * Create the kernel thread and wait for it to start. - */ - nlmsvc_rqst = svc_prepare_thread(serv, &serv->sv_pools[0], NUMA_NO_NODE); - if (IS_ERR(nlmsvc_rqst)) { - error = PTR_ERR(nlmsvc_rqst); - printk(KERN_WARNING - "lockd_up: svc_rqst allocation failed, error=%d\n", - error); - lockd_unregister_notifiers(); - goto out_rqst; - } - - atomic_inc(&nlm_ntf_refcnt); - svc_sock_update_bufs(serv); - serv->sv_maxconn = nlm_max_connections; - - nlmsvc_task = kthread_create(lockd, nlmsvc_rqst, "%s", serv->sv_name); - if (IS_ERR(nlmsvc_task)) { - error = PTR_ERR(nlmsvc_task); - printk(KERN_WARNING - "lockd_up: kthread_run failed, error=%d\n", error); - goto out_task; - } - nlmsvc_rqst->rq_task = nlmsvc_task; - wake_up_process(nlmsvc_task); - - dprintk("lockd_up: service started\n"); - return 0; - -out_task: - lockd_svc_exit_thread(); - nlmsvc_task = NULL; -out_rqst: - nlmsvc_rqst = NULL; - return error; -} - static const struct svc_serv_ops lockd_sv_ops = { .svo_shutdown = svc_rpcb_cleanup, + .svo_function = lockd, .svo_enqueue_xprt = svc_xprt_do_enqueue, + .svo_module = THIS_MODULE, }; -static struct svc_serv *lockd_create_svc(void) +static int lockd_get(void) { struct svc_serv *serv; + int error; - /* - * Check whether we're already up and running. - */ - if (nlmsvc_rqst) { - /* - * Note: increase service usage, because later in case of error - * svc_destroy() will be called. - */ - svc_get(nlmsvc_rqst->rq_server); - return nlmsvc_rqst->rq_server; + if (nlmsvc_serv) { + svc_get(nlmsvc_serv); + nlmsvc_users++; + return 0; } /* @@ -454,14 +383,41 @@ static struct svc_serv *lockd_create_svc(void) serv = svc_create(&nlmsvc_program, LOCKD_BUFSIZE, &lockd_sv_ops); if (!serv) { printk(KERN_WARNING "lockd_up: create service failed\n"); - return ERR_PTR(-ENOMEM); + return -ENOMEM; } + + serv->sv_maxconn = nlm_max_connections; + error = svc_set_num_threads(serv, NULL, 1); + /* The thread now holds the only reference */ + svc_put(serv); + if (error < 0) + return error; + + nlmsvc_serv = serv; register_inetaddr_notifier(&lockd_inetaddr_notifier); #if IS_ENABLED(CONFIG_IPV6) register_inet6addr_notifier(&lockd_inet6addr_notifier); #endif dprintk("lockd_up: service created\n"); - return serv; + nlmsvc_users++; + return 0; +} + +static void lockd_put(void) +{ + if (WARN(nlmsvc_users <= 0, "lockd_down: no users!\n")) + return; + if (--nlmsvc_users) + return; + + unregister_inetaddr_notifier(&lockd_inetaddr_notifier); +#if IS_ENABLED(CONFIG_IPV6) + unregister_inet6addr_notifier(&lockd_inet6addr_notifier); +#endif + + svc_set_num_threads(nlmsvc_serv, NULL, 0); + nlmsvc_serv = NULL; + dprintk("lockd_down: service destroyed\n"); } /* @@ -469,36 +425,21 @@ static struct svc_serv *lockd_create_svc(void) */ int lockd_up(struct net *net, const struct cred *cred) { - struct svc_serv *serv; int error; mutex_lock(&nlmsvc_mutex); - serv = lockd_create_svc(); - if (IS_ERR(serv)) { - error = PTR_ERR(serv); - goto err_create; - } + error = lockd_get(); + if (error) + goto err; - error = lockd_up_net(serv, net, cred); + error = lockd_up_net(nlmsvc_serv, net, cred); if (error < 0) { - lockd_unregister_notifiers(); - goto err_put; + lockd_put(); + goto err; } - error = lockd_start_svc(serv); - if (error < 0) { - lockd_down_net(serv, net); - goto err_put; - } - nlmsvc_users++; - /* - * Note: svc_serv structures have an initial use count of 1, - * so we exit through here on both success and failure. - */ -err_put: - svc_destroy(serv); -err_create: +err: mutex_unlock(&nlmsvc_mutex); return error; } @@ -511,27 +452,8 @@ void lockd_down(struct net *net) { mutex_lock(&nlmsvc_mutex); - lockd_down_net(nlmsvc_rqst->rq_server, net); - if (nlmsvc_users) { - if (--nlmsvc_users) - goto out; - } else { - printk(KERN_ERR "lockd_down: no users! task=%p\n", - nlmsvc_task); - BUG(); - } - - if (!nlmsvc_task) { - printk(KERN_ERR "lockd_down: no lockd running.\n"); - BUG(); - } - kthread_stop(nlmsvc_task); - dprintk("lockd_down: service stopped\n"); - lockd_svc_exit_thread(); - dprintk("lockd_down: service destroyed\n"); - nlmsvc_task = NULL; - nlmsvc_rqst = NULL; -out: + lockd_down_net(nlmsvc_serv, net); + lockd_put(); mutex_unlock(&nlmsvc_mutex); } EXPORT_SYMBOL_GPL(lockd_down); diff --git a/fs/lockd/svclock.c b/fs/lockd/svclock.c index e9b85d8fd5fe..cb3658ab9b7a 100644 --- a/fs/lockd/svclock.c +++ b/fs/lockd/svclock.c @@ -470,8 +470,10 @@ nlmsvc_lock(struct svc_rqst *rqstp, struct nlm_file *file, struct nlm_host *host, struct nlm_lock *lock, int wait, struct nlm_cookie *cookie, int reclaim) { - struct nlm_block *block = NULL; +#if IS_ENABLED(CONFIG_SUNRPC_DEBUG) struct inode *inode = nlmsvc_file_inode(file); +#endif + struct nlm_block *block = NULL; int error; int mode; int async_block = 0; @@ -484,7 +486,7 @@ nlmsvc_lock(struct svc_rqst *rqstp, struct nlm_file *file, (long long)lock->fl.fl_end, wait); - if (inode->i_sb->s_export_op->flags & EXPORT_OP_SYNC_LOCKS) { + if (nlmsvc_file_file(file)->f_op->lock) { async_block = wait; wait = 0; } diff --git a/fs/locks.c b/fs/locks.c index 0fca9d680978..8c6df10cd9ed 100644 --- a/fs/locks.c +++ b/fs/locks.c @@ -62,6 +62,7 @@ #include <linux/pid_namespace.h> #include <linux/hashtable.h> #include <linux/percpu.h> +#include <linux/sysctl.h> #define CREATE_TRACE_POINTS #include <trace/events/filelock.h> @@ -88,8 +89,37 @@ static int target_leasetype(struct file_lock *fl) return fl->fl_type; } -int leases_enable = 1; -int lease_break_time = 45; +static int leases_enable = 1; +static int lease_break_time = 45; + +#ifdef CONFIG_SYSCTL +static struct ctl_table locks_sysctls[] = { + { + .procname = "leases-enable", + .data = &leases_enable, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec, + }, +#ifdef CONFIG_MMU + { + .procname = "lease-break-time", + .data = &lease_break_time, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec, + }, +#endif /* CONFIG_MMU */ + {} +}; + +static int __init init_fs_locks_sysctls(void) +{ + register_sysctl_init("fs", locks_sysctls); + return 0; +} +early_initcall(init_fs_locks_sysctls); +#endif /* CONFIG_SYSCTL */ /* * The global file_lock_list is only used for displaying /proc/locks, so we diff --git a/fs/mpage.c b/fs/mpage.c index 334e7d09aa65..87f5cfef6caa 100644 --- a/fs/mpage.c +++ b/fs/mpage.c @@ -29,7 +29,6 @@ #include <linux/writeback.h> #include <linux/backing-dev.h> #include <linux/pagevec.h> -#include <linux/cleancache.h> #include "internal.h" /* @@ -284,12 +283,6 @@ static struct bio *do_mpage_readpage(struct mpage_readpage_args *args) SetPageMappedToDisk(page); } - if (fully_mapped && blocks_per_page == 1 && !PageUptodate(page) && - cleancache_get_page(page) == 0) { - SetPageUptodate(page); - goto confused; - } - /* * This page will go to BIO. Do we need to send this BIO off first? */ diff --git a/fs/namei.c b/fs/namei.c index d81f04f8d818..b867a92c078e 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -1020,10 +1020,60 @@ static inline void put_link(struct nameidata *nd) path_put(&last->link); } -int sysctl_protected_symlinks __read_mostly = 0; -int sysctl_protected_hardlinks __read_mostly = 0; -int sysctl_protected_fifos __read_mostly; -int sysctl_protected_regular __read_mostly; +static int sysctl_protected_symlinks __read_mostly; +static int sysctl_protected_hardlinks __read_mostly; +static int sysctl_protected_fifos __read_mostly; +static int sysctl_protected_regular __read_mostly; + +#ifdef CONFIG_SYSCTL +static struct ctl_table namei_sysctls[] = { + { + .procname = "protected_symlinks", + .data = &sysctl_protected_symlinks, + .maxlen = sizeof(int), + .mode = 0600, + .proc_handler = proc_dointvec_minmax, + .extra1 = SYSCTL_ZERO, + .extra2 = SYSCTL_ONE, + }, + { + .procname = "protected_hardlinks", + .data = &sysctl_protected_hardlinks, + .maxlen = sizeof(int), + .mode = 0600, + .proc_handler = proc_dointvec_minmax, + .extra1 = SYSCTL_ZERO, + .extra2 = SYSCTL_ONE, + }, + { + .procname = "protected_fifos", + .data = &sysctl_protected_fifos, + .maxlen = sizeof(int), + .mode = 0600, + .proc_handler = proc_dointvec_minmax, + .extra1 = SYSCTL_ZERO, + .extra2 = SYSCTL_TWO, + }, + { + .procname = "protected_regular", + .data = &sysctl_protected_regular, + .maxlen = sizeof(int), + .mode = 0600, + .proc_handler = proc_dointvec_minmax, + .extra1 = SYSCTL_ZERO, + .extra2 = SYSCTL_TWO, + }, + { } +}; + +static int __init init_fs_namei_sysctls(void) +{ + register_sysctl_init("fs", namei_sysctls); + return 0; +} +fs_initcall(init_fs_namei_sysctls); + +#endif /* CONFIG_SYSCTL */ /** * may_follow_link - Check symlink following for unsafe situations diff --git a/fs/namespace.c b/fs/namespace.c index dc31ad6b370f..40b994a29e90 100644 --- a/fs/namespace.c +++ b/fs/namespace.c @@ -37,7 +37,7 @@ #include "internal.h" /* Maximum number of mounts in a mount namespace */ -unsigned int sysctl_mount_max __read_mostly = 100000; +static unsigned int sysctl_mount_max __read_mostly = 100000; static unsigned int m_hash_mask __read_mostly; static unsigned int m_hash_shift __read_mostly; @@ -4620,3 +4620,25 @@ const struct proc_ns_operations mntns_operations = { .install = mntns_install, .owner = mntns_owner, }; + +#ifdef CONFIG_SYSCTL +static struct ctl_table fs_namespace_sysctls[] = { + { + .procname = "mount-max", + .data = &sysctl_mount_max, + .maxlen = sizeof(unsigned int), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = SYSCTL_ONE, + }, + { } +}; + +static int __init init_fs_namespace_sysctls(void) +{ + register_sysctl_init("fs", fs_namespace_sysctls); + return 0; +} +fs_initcall(init_fs_namespace_sysctls); + +#endif /* CONFIG_SYSCTL */ diff --git a/fs/netfs/read_helper.c b/fs/netfs/read_helper.c index 6169659857b3..501da990c259 100644 --- a/fs/netfs/read_helper.c +++ b/fs/netfs/read_helper.c @@ -55,7 +55,8 @@ static struct netfs_read_request *netfs_alloc_read_request( INIT_WORK(&rreq->work, netfs_rreq_work); refcount_set(&rreq->usage, 1); __set_bit(NETFS_RREQ_IN_PROGRESS, &rreq->flags); - ops->init_rreq(rreq, file); + if (ops->init_rreq) + ops->init_rreq(rreq, file); netfs_stat(&netfs_n_rh_rreq); } diff --git a/fs/nfs/callback.c b/fs/nfs/callback.c index 86d856de1389..054cc1255fac 100644 --- a/fs/nfs/callback.c +++ b/fs/nfs/callback.c @@ -93,7 +93,7 @@ nfs4_callback_svc(void *vrqstp) svc_process(rqstp); } svc_exit_thread(rqstp); - module_put_and_exit(0); + module_put_and_kthread_exit(0); return 0; } @@ -137,7 +137,7 @@ nfs41_callback_svc(void *vrqstp) } } svc_exit_thread(rqstp); - module_put_and_exit(0); + module_put_and_kthread_exit(0); return 0; } @@ -169,12 +169,12 @@ static int nfs_callback_start_svc(int minorversion, struct rpc_xprt *xprt, if (nrservs < NFS4_MIN_NR_CALLBACK_THREADS) nrservs = NFS4_MIN_NR_CALLBACK_THREADS; - if (serv->sv_nrthreads-1 == nrservs) + if (serv->sv_nrthreads == nrservs) return 0; - ret = serv->sv_ops->svo_setup(serv, NULL, nrservs); + ret = svc_set_num_threads(serv, NULL, nrservs); if (ret) { - serv->sv_ops->svo_setup(serv, NULL, 0); + svc_set_num_threads(serv, NULL, 0); return ret; } dprintk("nfs_callback_up: service started\n"); @@ -235,14 +235,12 @@ err_bind: static const struct svc_serv_ops nfs40_cb_sv_ops = { .svo_function = nfs4_callback_svc, .svo_enqueue_xprt = svc_xprt_do_enqueue, - .svo_setup = svc_set_num_threads_sync, .svo_module = THIS_MODULE, }; #if defined(CONFIG_NFS_V4_1) static const struct svc_serv_ops nfs41_cb_sv_ops = { .svo_function = nfs41_callback_svc, .svo_enqueue_xprt = svc_xprt_do_enqueue, - .svo_setup = svc_set_num_threads_sync, .svo_module = THIS_MODULE, }; @@ -266,14 +264,8 @@ static struct svc_serv *nfs_callback_create_svc(int minorversion) /* * Check whether we're already up and running. */ - if (cb_info->serv) { - /* - * Note: increase service usage, because later in case of error - * svc_destroy() will be called. - */ - svc_get(cb_info->serv); - return cb_info->serv; - } + if (cb_info->serv) + return svc_get(cb_info->serv); switch (minorversion) { case 0: @@ -294,7 +286,7 @@ static struct svc_serv *nfs_callback_create_svc(int minorversion) printk(KERN_WARNING "nfs_callback_create_svc: no kthread, %d users??\n", cb_info->users); - serv = svc_create_pooled(&nfs4_callback_program, NFS4_CALLBACK_BUFSIZE, sv_ops); + serv = svc_create(&nfs4_callback_program, NFS4_CALLBACK_BUFSIZE, sv_ops); if (!serv) { printk(KERN_ERR "nfs_callback_create_svc: create service failed\n"); return ERR_PTR(-ENOMEM); @@ -335,16 +327,10 @@ int nfs_callback_up(u32 minorversion, struct rpc_xprt *xprt) goto err_start; cb_info->users++; - /* - * svc_create creates the svc_serv with sv_nrthreads == 1, and then - * svc_prepare_thread increments that. So we need to call svc_destroy - * on both success and failure so that the refcount is 1 when the - * thread exits. - */ err_net: if (!cb_info->users) cb_info->serv = NULL; - svc_destroy(serv); + svc_put(serv); err_create: mutex_unlock(&nfs_callback_mutex); return ret; @@ -369,8 +355,8 @@ void nfs_callback_down(int minorversion, struct net *net) cb_info->users--; if (cb_info->users == 0) { svc_get(serv); - serv->sv_ops->svo_setup(serv, NULL, 0); - svc_destroy(serv); + svc_set_num_threads(serv, NULL, 0); + svc_put(serv); dprintk("nfs_callback_down: service destroyed\n"); cb_info->serv = NULL; } diff --git a/fs/nfs/export.c b/fs/nfs/export.c index 171c424cb6d5..01596f2d0a1e 100644 --- a/fs/nfs/export.c +++ b/fs/nfs/export.c @@ -158,5 +158,5 @@ const struct export_operations nfs_export_ops = { .fetch_iversion = nfs_fetch_iversion, .flags = EXPORT_OP_NOWCC|EXPORT_OP_NOSUBTREECHK| EXPORT_OP_CLOSE_BEFORE_UNLINK|EXPORT_OP_REMOTE_FS| - EXPORT_OP_NOATOMIC_ATTR|EXPORT_OP_SYNC_LOCKS, + EXPORT_OP_NOATOMIC_ATTR, }; diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c index f63dfa01001c..d88b779f9dd0 100644 --- a/fs/nfs/nfs4state.c +++ b/fs/nfs/nfs4state.c @@ -2693,6 +2693,6 @@ static int nfs4_run_state_manager(void *ptr) allow_signal(SIGKILL); nfs4_state_manager(clp); nfs_put_client(clp); - module_put_and_exit(0); + module_put_and_kthread_exit(0); return 0; } diff --git a/fs/nfsd/filecache.c b/fs/nfsd/filecache.c index fdf89fcf1a0c..8bc807c5fea4 100644 --- a/fs/nfsd/filecache.c +++ b/fs/nfsd/filecache.c @@ -44,12 +44,9 @@ struct nfsd_fcache_bucket { static DEFINE_PER_CPU(unsigned long, nfsd_file_cache_hits); struct nfsd_fcache_disposal { - struct list_head list; struct work_struct work; - struct net *net; spinlock_t lock; struct list_head freeme; - struct rcu_head rcu; }; static struct workqueue_struct *nfsd_filecache_wq __read_mostly; @@ -62,8 +59,6 @@ static long nfsd_file_lru_flags; static struct fsnotify_group *nfsd_file_fsnotify_group; static atomic_long_t nfsd_filecache_count; static struct delayed_work nfsd_filecache_laundrette; -static DEFINE_SPINLOCK(laundrette_lock); -static LIST_HEAD(laundrettes); static void nfsd_file_gc(void); @@ -194,7 +189,6 @@ nfsd_file_alloc(struct inode *inode, unsigned int may, unsigned int hashval, __set_bit(NFSD_FILE_BREAK_READ, &nf->nf_flags); } nf->nf_mark = NULL; - init_rwsem(&nf->nf_rwsem); trace_nfsd_file_alloc(nf); } return nf; @@ -249,7 +243,7 @@ nfsd_file_do_unhash(struct nfsd_file *nf) trace_nfsd_file_unhash(nf); if (nfsd_file_check_write_error(nf)) - nfsd_reset_boot_verifier(net_generic(nf->nf_net, nfsd_net_id)); + nfsd_reset_write_verifier(net_generic(nf->nf_net, nfsd_net_id)); --nfsd_file_hashtbl[nf->nf_hashval].nfb_count; hlist_del_rcu(&nf->nf_node); atomic_long_dec(&nfsd_filecache_count); @@ -367,19 +361,13 @@ nfsd_file_list_remove_disposal(struct list_head *dst, static void nfsd_file_list_add_disposal(struct list_head *files, struct net *net) { - struct nfsd_fcache_disposal *l; + struct nfsd_net *nn = net_generic(net, nfsd_net_id); + struct nfsd_fcache_disposal *l = nn->fcache_disposal; - rcu_read_lock(); - list_for_each_entry_rcu(l, &laundrettes, list) { - if (l->net == net) { - spin_lock(&l->lock); - list_splice_tail_init(files, &l->freeme); - spin_unlock(&l->lock); - queue_work(nfsd_filecache_wq, &l->work); - break; - } - } - rcu_read_unlock(); + spin_lock(&l->lock); + list_splice_tail_init(files, &l->freeme); + spin_unlock(&l->lock); + queue_work(nfsd_filecache_wq, &l->work); } static void @@ -755,7 +743,7 @@ nfsd_file_cache_purge(struct net *net) } static struct nfsd_fcache_disposal * -nfsd_alloc_fcache_disposal(struct net *net) +nfsd_alloc_fcache_disposal(void) { struct nfsd_fcache_disposal *l; @@ -763,7 +751,6 @@ nfsd_alloc_fcache_disposal(struct net *net) if (!l) return NULL; INIT_WORK(&l->work, nfsd_file_delayed_close); - l->net = net; spin_lock_init(&l->lock); INIT_LIST_HEAD(&l->freeme); return l; @@ -772,61 +759,27 @@ nfsd_alloc_fcache_disposal(struct net *net) static void nfsd_free_fcache_disposal(struct nfsd_fcache_disposal *l) { - rcu_assign_pointer(l->net, NULL); cancel_work_sync(&l->work); nfsd_file_dispose_list(&l->freeme); - kfree_rcu(l, rcu); -} - -static void -nfsd_add_fcache_disposal(struct nfsd_fcache_disposal *l) -{ - spin_lock(&laundrette_lock); - list_add_tail_rcu(&l->list, &laundrettes); - spin_unlock(&laundrette_lock); -} - -static void -nfsd_del_fcache_disposal(struct nfsd_fcache_disposal *l) -{ - spin_lock(&laundrette_lock); - list_del_rcu(&l->list); - spin_unlock(&laundrette_lock); -} - -static int -nfsd_alloc_fcache_disposal_net(struct net *net) -{ - struct nfsd_fcache_disposal *l; - - l = nfsd_alloc_fcache_disposal(net); - if (!l) - return -ENOMEM; - nfsd_add_fcache_disposal(l); - return 0; + kfree(l); } static void nfsd_free_fcache_disposal_net(struct net *net) { - struct nfsd_fcache_disposal *l; + struct nfsd_net *nn = net_generic(net, nfsd_net_id); + struct nfsd_fcache_disposal *l = nn->fcache_disposal; - rcu_read_lock(); - list_for_each_entry_rcu(l, &laundrettes, list) { - if (l->net != net) - continue; - nfsd_del_fcache_disposal(l); - rcu_read_unlock(); - nfsd_free_fcache_disposal(l); - return; - } - rcu_read_unlock(); + nfsd_free_fcache_disposal(l); } int nfsd_file_cache_start_net(struct net *net) { - return nfsd_alloc_fcache_disposal_net(net); + struct nfsd_net *nn = net_generic(net, nfsd_net_id); + + nn->fcache_disposal = nfsd_alloc_fcache_disposal(); + return nn->fcache_disposal ? 0 : -ENOMEM; } void diff --git a/fs/nfsd/filecache.h b/fs/nfsd/filecache.h index 7872df5a0fe3..435ceab27897 100644 --- a/fs/nfsd/filecache.h +++ b/fs/nfsd/filecache.h @@ -46,7 +46,6 @@ struct nfsd_file { refcount_t nf_ref; unsigned char nf_may; struct nfsd_file_mark *nf_mark; - struct rw_semaphore nf_rwsem; }; int nfsd_file_cache_init(void); diff --git a/fs/nfsd/netns.h b/fs/nfsd/netns.h index 935c1028c217..1b1a962a1804 100644 --- a/fs/nfsd/netns.h +++ b/fs/nfsd/netns.h @@ -11,6 +11,7 @@ #include <net/net_namespace.h> #include <net/netns/generic.h> #include <linux/percpu_counter.h> +#include <linux/siphash.h> /* Hash tables for nfs4_clientid state */ #define CLIENT_HASH_BITS 4 @@ -108,9 +109,8 @@ struct nfsd_net { bool nfsd_net_up; bool lockd_up; - /* Time of server startup */ - struct timespec64 nfssvc_boot; - seqlock_t boot_lock; + seqlock_t writeverf_lock; + unsigned char writeverf[8]; /* * Max number of connections this nfsd container will allow. Defaults @@ -123,12 +123,13 @@ struct nfsd_net { u32 clverifier_counter; struct svc_serv *nfsd_serv; - - wait_queue_head_t ntf_wq; - atomic_t ntf_refcnt; - - /* Allow umount to wait for nfsd state cleanup */ - struct completion nfsd_shutdown_complete; + /* When a listening socket is added to nfsd, keep_active is set + * and this justifies a reference on nfsd_serv. This stops + * nfsd_serv from being freed. When the number of threads is + * set, keep_active is cleared and the reference is dropped. So + * when the last thread exits, the service will be destroyed. + */ + int keep_active; /* * clientid and stateid data for construction of net unique COPY @@ -184,6 +185,10 @@ struct nfsd_net { /* utsname taken from the process that starts the server */ char nfsd_name[UNX_MAXNODENAME+1]; + + struct nfsd_fcache_disposal *fcache_disposal; + + siphash_key_t siphash_key; }; /* Simple check to find out if a given net was properly initialized */ @@ -193,6 +198,6 @@ extern void nfsd_netns_free_versions(struct nfsd_net *nn); extern unsigned int nfsd_net_id; -void nfsd_copy_boot_verifier(__be32 verf[2], struct nfsd_net *nn); -void nfsd_reset_boot_verifier(struct nfsd_net *nn); +void nfsd_copy_write_verifier(__be32 verf[2], struct nfsd_net *nn); +void nfsd_reset_write_verifier(struct nfsd_net *nn); #endif /* __NFSD_NETNS_H__ */ diff --git a/fs/nfsd/nfs3proc.c b/fs/nfsd/nfs3proc.c index 15dac36ca852..8ef53f6726ec 100644 --- a/fs/nfsd/nfs3proc.c +++ b/fs/nfsd/nfs3proc.c @@ -202,15 +202,11 @@ nfsd3_proc_write(struct svc_rqst *rqstp) fh_copy(&resp->fh, &argp->fh); resp->committed = argp->stable; nvecs = svc_fill_write_vector(rqstp, &argp->payload); - if (!nvecs) { - resp->status = nfserr_io; - goto out; - } + resp->status = nfsd_write(rqstp, &resp->fh, argp->offset, rqstp->rq_vec, nvecs, &cnt, resp->committed, resp->verf); resp->count = cnt; -out: return rpc_success; } diff --git a/fs/nfsd/nfs3xdr.c b/fs/nfsd/nfs3xdr.c index c3ac1b6aa3aa..7c45ba4db61b 100644 --- a/fs/nfsd/nfs3xdr.c +++ b/fs/nfsd/nfs3xdr.c @@ -487,71 +487,6 @@ neither: return true; } -static bool fs_supports_change_attribute(struct super_block *sb) -{ - return sb->s_flags & SB_I_VERSION || sb->s_export_op->fetch_iversion; -} - -/* - * Fill in the pre_op attr for the wcc data - */ -void fill_pre_wcc(struct svc_fh *fhp) -{ - struct inode *inode; - struct kstat stat; - bool v4 = (fhp->fh_maxsize == NFS4_FHSIZE); - - if (fhp->fh_no_wcc || fhp->fh_pre_saved) - return; - inode = d_inode(fhp->fh_dentry); - if (fs_supports_change_attribute(inode->i_sb) || !v4) { - __be32 err = fh_getattr(fhp, &stat); - - if (err) { - /* Grab the times from inode anyway */ - stat.mtime = inode->i_mtime; - stat.ctime = inode->i_ctime; - stat.size = inode->i_size; - } - fhp->fh_pre_mtime = stat.mtime; - fhp->fh_pre_ctime = stat.ctime; - fhp->fh_pre_size = stat.size; - } - if (v4) - fhp->fh_pre_change = nfsd4_change_attribute(&stat, inode); - - fhp->fh_pre_saved = true; -} - -/* - * Fill in the post_op attr for the wcc data - */ -void fill_post_wcc(struct svc_fh *fhp) -{ - bool v4 = (fhp->fh_maxsize == NFS4_FHSIZE); - struct inode *inode = d_inode(fhp->fh_dentry); - - if (fhp->fh_no_wcc) - return; - - if (fhp->fh_post_saved) - printk("nfsd: inode locked twice during operation.\n"); - - fhp->fh_post_saved = true; - - if (fs_supports_change_attribute(inode->i_sb) || !v4) { - __be32 err = fh_getattr(fhp, &fhp->fh_post_attr); - - if (err) { - fhp->fh_post_saved = false; - fhp->fh_post_attr.ctime = inode->i_ctime; - } - } - if (v4) - fhp->fh_post_change = - nfsd4_change_attribute(&fhp->fh_post_attr, inode); -} - /* * XDR decode functions */ diff --git a/fs/nfsd/nfs4proc.c b/fs/nfsd/nfs4proc.c index a36261f89bdf..ed1ee25647be 100644 --- a/fs/nfsd/nfs4proc.c +++ b/fs/nfsd/nfs4proc.c @@ -598,7 +598,7 @@ static void gen_boot_verifier(nfs4_verifier *verifier, struct net *net) BUILD_BUG_ON(2*sizeof(*verf) != sizeof(verifier->data)); - nfsd_copy_boot_verifier(verf, net_generic(net, nfsd_net_id)); + nfsd_copy_write_verifier(verf, net_generic(net, nfsd_net_id)); } static __be32 @@ -1101,7 +1101,7 @@ nfsd4_clone(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, if (status) goto out; - status = nfsd4_clone_file_range(src, clone->cl_src_pos, + status = nfsd4_clone_file_range(rqstp, src, clone->cl_src_pos, dst, clone->cl_dst_pos, clone->cl_count, EX_ISSYNC(cstate->current_fh.fh_export)); @@ -1510,11 +1510,14 @@ static void nfsd4_init_copy_res(struct nfsd4_copy *copy, bool sync) static ssize_t _nfsd_copy_file_range(struct nfsd4_copy *copy) { + struct file *dst = copy->nf_dst->nf_file; + struct file *src = copy->nf_src->nf_file; + errseq_t since; ssize_t bytes_copied = 0; u64 bytes_total = copy->cp_count; u64 src_pos = copy->cp_src_pos; u64 dst_pos = copy->cp_dst_pos; - __be32 status; + int status; /* See RFC 7862 p.67: */ if (bytes_total == 0) @@ -1522,9 +1525,8 @@ static ssize_t _nfsd_copy_file_range(struct nfsd4_copy *copy) do { if (kthread_should_stop()) break; - bytes_copied = nfsd_copy_file_range(copy->nf_src->nf_file, - src_pos, copy->nf_dst->nf_file, dst_pos, - bytes_total); + bytes_copied = nfsd_copy_file_range(src, src_pos, dst, dst_pos, + bytes_total); if (bytes_copied <= 0) break; bytes_total -= bytes_copied; @@ -1534,11 +1536,11 @@ static ssize_t _nfsd_copy_file_range(struct nfsd4_copy *copy) } while (bytes_total > 0 && !copy->cp_synchronous); /* for a non-zero asynchronous copy do a commit of data */ if (!copy->cp_synchronous && copy->cp_res.wr_bytes_written > 0) { - down_write(©->nf_dst->nf_rwsem); - status = vfs_fsync_range(copy->nf_dst->nf_file, - copy->cp_dst_pos, + since = READ_ONCE(dst->f_wb_err); + status = vfs_fsync_range(dst, copy->cp_dst_pos, copy->cp_res.wr_bytes_written, 0); - up_write(©->nf_dst->nf_rwsem); + if (!status) + status = filemap_check_wb_err(dst->f_mapping, since); if (!status) copy->committed = true; } @@ -2528,7 +2530,7 @@ nfsd4_proc_compound(struct svc_rqst *rqstp) goto encode_op; } - fh_clear_wcc(current_fh); + fh_clear_pre_post_attrs(current_fh); /* If op is non-idempotent */ if (op->opdesc->op_flags & OP_MODIFIES_SOMETHING) { diff --git a/fs/nfsd/nfs4state.c b/fs/nfsd/nfs4state.c index 1956d377d1a6..72900b89cf84 100644 --- a/fs/nfsd/nfs4state.c +++ b/fs/nfsd/nfs4state.c @@ -246,6 +246,7 @@ find_blocked_lock(struct nfs4_lockowner *lo, struct knfsd_fh *fh, list_for_each_entry(cur, &lo->lo_blocked, nbl_list) { if (fh_match(fh, &cur->nbl_fh)) { list_del_init(&cur->nbl_list); + WARN_ON(list_empty(&cur->nbl_lru)); list_del_init(&cur->nbl_lru); found = cur; break; @@ -271,6 +272,7 @@ find_or_allocate_block(struct nfs4_lockowner *lo, struct knfsd_fh *fh, INIT_LIST_HEAD(&nbl->nbl_lru); fh_copy_shallow(&nbl->nbl_fh, fh); locks_init_lock(&nbl->nbl_lock); + kref_init(&nbl->nbl_kref); nfsd4_init_cb(&nbl->nbl_cb, lo->lo_owner.so_client, &nfsd4_cb_notify_lock_ops, NFSPROC4_CLNT_CB_NOTIFY_LOCK); @@ -280,11 +282,20 @@ find_or_allocate_block(struct nfs4_lockowner *lo, struct knfsd_fh *fh, } static void +free_nbl(struct kref *kref) +{ + struct nfsd4_blocked_lock *nbl; + + nbl = container_of(kref, struct nfsd4_blocked_lock, nbl_kref); + kfree(nbl); +} + +static void free_blocked_lock(struct nfsd4_blocked_lock *nbl) { locks_delete_block(&nbl->nbl_lock); locks_release_private(&nbl->nbl_lock); - kfree(nbl); + kref_put(&nbl->nbl_kref, free_nbl); } static void @@ -302,6 +313,7 @@ remove_blocked_locks(struct nfs4_lockowner *lo) struct nfsd4_blocked_lock, nbl_list); list_del_init(&nbl->nbl_list); + WARN_ON(list_empty(&nbl->nbl_lru)); list_move(&nbl->nbl_lru, &reaplist); } spin_unlock(&nn->blocked_locks_lock); @@ -360,11 +372,13 @@ static const struct nfsd4_callback_ops nfsd4_cb_notify_lock_ops = { * st_{access,deny}_bmap field of the stateid, in order to track not * only what share bits are currently in force, but also what * combinations of share bits previous opens have used. This allows us - * to enforce the recommendation of rfc 3530 14.2.19 that the server - * return an error if the client attempt to downgrade to a combination - * of share bits not explicable by closing some of its previous opens. + * to enforce the recommendation in + * https://datatracker.ietf.org/doc/html/rfc7530#section-16.19.4 that + * the server return an error if the client attempt to downgrade to a + * combination of share bits not explicable by closing some of its + * previous opens. * - * XXX: This enforcement is actually incomplete, since we don't keep + * This enforcement is arguably incomplete, since we don't keep * track of access/deny bit combinations; so, e.g., we allow: * * OPEN allow read, deny write @@ -372,6 +386,10 @@ static const struct nfsd4_callback_ops nfsd4_cb_notify_lock_ops = { * DOWNGRADE allow read, deny none * * which we should reject. + * + * But you could also argue that our current code is already overkill, + * since it only exists to return NFS4ERR_INVAL on incorrect client + * behavior. */ static unsigned int bmap_to_share_mode(unsigned long bmap) @@ -6040,7 +6058,11 @@ nfs4_preprocess_stateid_op(struct svc_rqst *rqstp, *nfp = NULL; if (ZERO_STATEID(stateid) || ONE_STATEID(stateid)) { - status = check_special_stateids(net, fhp, stateid, flags); + if (cstid) + status = nfserr_bad_stateid; + else + status = check_special_stateids(net, fhp, stateid, + flags); goto done; } @@ -6836,7 +6858,6 @@ nfsd4_lock(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, struct nfsd4_blocked_lock *nbl = NULL; struct file_lock *file_lock = NULL; struct file_lock *conflock = NULL; - struct super_block *sb; __be32 status = 0; int lkflg; int err; @@ -6858,7 +6879,6 @@ nfsd4_lock(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, dprintk("NFSD: nfsd4_lock: permission denied!\n"); return status; } - sb = cstate->current_fh.fh_dentry->d_sb; if (lock->lk_is_new) { if (nfsd4_has_session(cstate)) @@ -6910,8 +6930,7 @@ nfsd4_lock(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, fp = lock_stp->st_stid.sc_file; switch (lock->lk_type) { case NFS4_READW_LT: - if (nfsd4_has_session(cstate) && - !(sb->s_export_op->flags & EXPORT_OP_SYNC_LOCKS)) + if (nfsd4_has_session(cstate)) fl_flags |= FL_SLEEP; fallthrough; case NFS4_READ_LT: @@ -6923,8 +6942,7 @@ nfsd4_lock(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, fl_type = F_RDLCK; break; case NFS4_WRITEW_LT: - if (nfsd4_has_session(cstate) && - !(sb->s_export_op->flags & EXPORT_OP_SYNC_LOCKS)) + if (nfsd4_has_session(cstate)) fl_flags |= FL_SLEEP; fallthrough; case NFS4_WRITE_LT: @@ -6945,6 +6963,16 @@ nfsd4_lock(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, goto out; } + /* + * Most filesystems with their own ->lock operations will block + * the nfsd thread waiting to acquire the lock. That leads to + * deadlocks (we don't want every nfsd thread tied up waiting + * for file locks), so don't attempt blocking lock notifications + * on those filesystems: + */ + if (nf->nf_file->f_op->lock) + fl_flags &= ~FL_SLEEP; + nbl = find_or_allocate_block(lock_sop, &fp->fi_fhandle, nn); if (!nbl) { dprintk("NFSD: %s: unable to allocate block!\n", __func__); @@ -6975,6 +7003,7 @@ nfsd4_lock(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, spin_lock(&nn->blocked_locks_lock); list_add_tail(&nbl->nbl_list, &lock_sop->lo_blocked); list_add_tail(&nbl->nbl_lru, &nn->blocked_locks_lru); + kref_get(&nbl->nbl_kref); spin_unlock(&nn->blocked_locks_lock); } @@ -6987,6 +7016,7 @@ nfsd4_lock(struct svc_rqst *rqstp, struct nfsd4_compound_state *cstate, nn->somebody_reclaimed = true; break; case FILE_LOCK_DEFERRED: + kref_put(&nbl->nbl_kref, free_nbl); nbl = NULL; fallthrough; case -EAGAIN: /* conflock holds conflicting lock */ @@ -7007,8 +7037,13 @@ out: /* dequeue it if we queued it before */ if (fl_flags & FL_SLEEP) { spin_lock(&nn->blocked_locks_lock); - list_del_init(&nbl->nbl_list); - list_del_init(&nbl->nbl_lru); + if (!list_empty(&nbl->nbl_list) && + !list_empty(&nbl->nbl_lru)) { + list_del_init(&nbl->nbl_list); + list_del_init(&nbl->nbl_lru); + kref_put(&nbl->nbl_kref, free_nbl); + } + /* nbl can use one of lists to be linked to reaplist */ spin_unlock(&nn->blocked_locks_lock); } free_blocked_lock(nbl); diff --git a/fs/nfsd/nfs4xdr.c b/fs/nfsd/nfs4xdr.c index 5a93a5db4fb0..899de438e529 100644 --- a/fs/nfsd/nfs4xdr.c +++ b/fs/nfsd/nfs4xdr.c @@ -277,21 +277,10 @@ nfsd4_decode_verifier4(struct nfsd4_compoundargs *argp, nfs4_verifier *verf) static __be32 nfsd4_decode_bitmap4(struct nfsd4_compoundargs *argp, u32 *bmval, u32 bmlen) { - u32 i, count; - __be32 *p; - - if (xdr_stream_decode_u32(argp->xdr, &count) < 0) - return nfserr_bad_xdr; - /* request sanity */ - if (count > 1000) - return nfserr_bad_xdr; - p = xdr_inline_decode(argp->xdr, count << 2); - if (!p) - return nfserr_bad_xdr; - for (i = 0; i < bmlen; i++) - bmval[i] = (i < count) ? be32_to_cpup(p++) : 0; + ssize_t status; - return nfs_ok; + status = xdr_stream_decode_uint32_array(argp->xdr, bmval, bmlen); + return status == -EBADMSG ? nfserr_bad_xdr : nfs_ok; } static __be32 @@ -4804,8 +4793,8 @@ nfsd4_encode_read_plus_hole(struct nfsd4_compoundres *resp, return nfserr_resource; *p++ = htonl(NFS4_CONTENT_HOLE); - p = xdr_encode_hyper(p, read->rd_offset); - p = xdr_encode_hyper(p, count); + p = xdr_encode_hyper(p, read->rd_offset); + p = xdr_encode_hyper(p, count); *eof = (read->rd_offset + count) >= f_size; *maxcount = min_t(unsigned long, count, *maxcount); diff --git a/fs/nfsd/nfscache.c b/fs/nfsd/nfscache.c index 6e0b6f3148dc..a4a69ab6ab28 100644 --- a/fs/nfsd/nfscache.c +++ b/fs/nfsd/nfscache.c @@ -87,7 +87,7 @@ nfsd_hashsize(unsigned int limit) static u32 nfsd_cache_hash(__be32 xid, struct nfsd_net *nn) { - return hash_32(be32_to_cpu(xid), nn->maskbits); + return hash_32((__force u32)xid, nn->maskbits); } static struct svc_cacherep * diff --git a/fs/nfsd/nfsctl.c b/fs/nfsd/nfsctl.c index 51a49e0cfe37..b9f27fbcd768 100644 --- a/fs/nfsd/nfsctl.c +++ b/fs/nfsd/nfsctl.c @@ -742,13 +742,12 @@ static ssize_t __write_ports_addfd(char *buf, struct net *net, const struct cred return err; err = svc_addsock(nn->nfsd_serv, fd, buf, SIMPLE_TRANSACTION_LIMIT, cred); - if (err < 0) { - nfsd_destroy(net); - return err; - } - /* Decrease the count, but don't shut down the service */ - nn->nfsd_serv->sv_nrthreads--; + if (err >= 0 && + !nn->nfsd_serv->sv_nrthreads && !xchg(&nn->keep_active, 1)) + svc_get(nn->nfsd_serv); + + nfsd_put(net); return err; } @@ -783,8 +782,10 @@ static ssize_t __write_ports_addxprt(char *buf, struct net *net, const struct cr if (err < 0 && err != -EAFNOSUPPORT) goto out_close; - /* Decrease the count, but don't shut down the service */ - nn->nfsd_serv->sv_nrthreads--; + if (!nn->nfsd_serv->sv_nrthreads && !xchg(&nn->keep_active, 1)) + svc_get(nn->nfsd_serv); + + nfsd_put(net); return 0; out_close: xprt = svc_find_xprt(nn->nfsd_serv, transport, net, PF_INET, port); @@ -793,10 +794,7 @@ out_close: svc_xprt_put(xprt); } out_err: - if (!list_empty(&nn->nfsd_serv->sv_permsocks)) - nn->nfsd_serv->sv_nrthreads--; - else - nfsd_destroy(net); + nfsd_put(net); return err; } @@ -1485,9 +1483,8 @@ static __net_init int nfsd_init_net(struct net *net) nn->clientid_counter = nn->clientid_base + 1; nn->s2s_cp_cl_id = nn->clientid_counter++; - atomic_set(&nn->ntf_refcnt, 0); - init_waitqueue_head(&nn->ntf_wq); - seqlock_init(&nn->boot_lock); + get_random_bytes(&nn->siphash_key, sizeof(nn->siphash_key)); + seqlock_init(&nn->writeverf_lock); return 0; diff --git a/fs/nfsd/nfsd.h b/fs/nfsd/nfsd.h index 498e5a489826..3e5008b475ff 100644 --- a/fs/nfsd/nfsd.h +++ b/fs/nfsd/nfsd.h @@ -97,7 +97,7 @@ int nfsd_pool_stats_open(struct inode *, struct file *); int nfsd_pool_stats_release(struct inode *, struct file *); void nfsd_shutdown_threads(struct net *net); -void nfsd_destroy(struct net *net); +void nfsd_put(struct net *net); bool i_am_nfsd(void); diff --git a/fs/nfsd/nfsfh.c b/fs/nfsd/nfsfh.c index f3779fa72c89..145208bcb9bd 100644 --- a/fs/nfsd/nfsfh.c +++ b/fs/nfsd/nfsfh.c @@ -611,6 +611,70 @@ out_negative: return nfserr_serverfault; } +#ifdef CONFIG_NFSD_V3 + +/** + * fh_fill_pre_attrs - Fill in pre-op attributes + * @fhp: file handle to be updated + * + */ +void fh_fill_pre_attrs(struct svc_fh *fhp) +{ + bool v4 = (fhp->fh_maxsize == NFS4_FHSIZE); + struct inode *inode; + struct kstat stat; + __be32 err; + + if (fhp->fh_no_wcc || fhp->fh_pre_saved) + return; + + inode = d_inode(fhp->fh_dentry); + err = fh_getattr(fhp, &stat); + if (err) { + /* Grab the times from inode anyway */ + stat.mtime = inode->i_mtime; + stat.ctime = inode->i_ctime; + stat.size = inode->i_size; + } + if (v4) + fhp->fh_pre_change = nfsd4_change_attribute(&stat, inode); + + fhp->fh_pre_mtime = stat.mtime; + fhp->fh_pre_ctime = stat.ctime; + fhp->fh_pre_size = stat.size; + fhp->fh_pre_saved = true; +} + +/** + * fh_fill_post_attrs - Fill in post-op attributes + * @fhp: file handle to be updated + * + */ +void fh_fill_post_attrs(struct svc_fh *fhp) +{ + bool v4 = (fhp->fh_maxsize == NFS4_FHSIZE); + struct inode *inode = d_inode(fhp->fh_dentry); + __be32 err; + + if (fhp->fh_no_wcc) + return; + + if (fhp->fh_post_saved) + printk("nfsd: inode locked twice during operation.\n"); + + err = fh_getattr(fhp, &fhp->fh_post_attr); + if (err) { + fhp->fh_post_saved = false; + fhp->fh_post_attr.ctime = inode->i_ctime; + } else + fhp->fh_post_saved = true; + if (v4) + fhp->fh_post_change = + nfsd4_change_attribute(&fhp->fh_post_attr, inode); +} + +#endif /* CONFIG_NFSD_V3 */ + /* * Release a file handle. */ @@ -623,7 +687,7 @@ fh_put(struct svc_fh *fhp) fh_unlock(fhp); fhp->fh_dentry = NULL; dput(dentry); - fh_clear_wcc(fhp); + fh_clear_pre_post_attrs(fhp); } fh_drop_write(fhp); if (exp) { diff --git a/fs/nfsd/nfsfh.h b/fs/nfsd/nfsfh.h index d11e4b6870d6..434930d8a946 100644 --- a/fs/nfsd/nfsfh.h +++ b/fs/nfsd/nfsfh.h @@ -284,12 +284,13 @@ static inline u32 knfsd_fh_hash(const struct knfsd_fh *fh) #endif #ifdef CONFIG_NFSD_V3 -/* - * The wcc data stored in current_fh should be cleared - * between compound ops. + +/** + * fh_clear_pre_post_attrs - Reset pre/post attributes + * @fhp: file handle to be updated + * */ -static inline void -fh_clear_wcc(struct svc_fh *fhp) +static inline void fh_clear_pre_post_attrs(struct svc_fh *fhp) { fhp->fh_post_saved = false; fhp->fh_pre_saved = false; @@ -323,13 +324,24 @@ static inline u64 nfsd4_change_attribute(struct kstat *stat, return time_to_chattr(&stat->ctime); } -extern void fill_pre_wcc(struct svc_fh *fhp); -extern void fill_post_wcc(struct svc_fh *fhp); -#else -#define fh_clear_wcc(ignored) -#define fill_pre_wcc(ignored) -#define fill_post_wcc(notused) -#endif /* CONFIG_NFSD_V3 */ +extern void fh_fill_pre_attrs(struct svc_fh *fhp); +extern void fh_fill_post_attrs(struct svc_fh *fhp); + +#else /* !CONFIG_NFSD_V3 */ + +static inline void fh_clear_pre_post_attrs(struct svc_fh *fhp) +{ +} + +static inline void fh_fill_pre_attrs(struct svc_fh *fhp) +{ +} + +static inline void fh_fill_post_attrs(struct svc_fh *fhp) +{ +} + +#endif /* !CONFIG_NFSD_V3 */ /* @@ -355,7 +367,7 @@ fh_lock_nested(struct svc_fh *fhp, unsigned int subclass) inode = d_inode(dentry); inode_lock_nested(inode, subclass); - fill_pre_wcc(fhp); + fh_fill_pre_attrs(fhp); fhp->fh_locked = true; } @@ -372,7 +384,7 @@ static inline void fh_unlock(struct svc_fh *fhp) { if (fhp->fh_locked) { - fill_post_wcc(fhp); + fh_fill_post_attrs(fhp); inode_unlock(d_inode(fhp->fh_dentry)); fhp->fh_locked = false; } diff --git a/fs/nfsd/nfsproc.c b/fs/nfsd/nfsproc.c index de282f3273c5..18b8eb43a19b 100644 --- a/fs/nfsd/nfsproc.c +++ b/fs/nfsd/nfsproc.c @@ -235,10 +235,6 @@ nfsd_proc_write(struct svc_rqst *rqstp) argp->len, argp->offset); nvecs = svc_fill_write_vector(rqstp, &argp->payload); - if (!nvecs) { - resp->status = nfserr_io; - goto out; - } resp->status = nfsd_write(rqstp, fh_copy(&resp->fh, &argp->fh), argp->offset, rqstp->rq_vec, nvecs, @@ -247,7 +243,6 @@ nfsd_proc_write(struct svc_rqst *rqstp) resp->status = fh_getattr(&resp->fh, &resp->stat); else if (resp->status == nfserr_jukebox) return rpc_drop_reply; -out: return rpc_success; } @@ -850,6 +845,7 @@ nfserrno (int errno) { nfserr_io, -EIO }, { nfserr_nxio, -ENXIO }, { nfserr_fbig, -E2BIG }, + { nfserr_stale, -EBADF }, { nfserr_acces, -EACCES }, { nfserr_exist, -EEXIST }, { nfserr_xdev, -EXDEV }, @@ -878,6 +874,8 @@ nfserrno (int errno) { nfserr_toosmall, -ETOOSMALL }, { nfserr_serverfault, -ESERVERFAULT }, { nfserr_serverfault, -ENFILE }, + { nfserr_io, -EREMOTEIO }, + { nfserr_stale, -EOPENSTALE }, { nfserr_io, -EUCLEAN }, { nfserr_perm, -ENOKEY }, { nfserr_no_grace, -ENOGRACE}, diff --git a/fs/nfsd/nfssvc.c b/fs/nfsd/nfssvc.c index 80431921e5d7..b8c682b62d29 100644 --- a/fs/nfsd/nfssvc.c +++ b/fs/nfsd/nfssvc.c @@ -12,6 +12,7 @@ #include <linux/module.h> #include <linux/fs_struct.h> #include <linux/swap.h> +#include <linux/siphash.h> #include <linux/sunrpc/stats.h> #include <linux/sunrpc/svcsock.h> @@ -55,18 +56,17 @@ static __be32 nfsd_init_request(struct svc_rqst *, struct svc_process_info *); /* - * nfsd_mutex protects nn->nfsd_serv -- both the pointer itself and the members - * of the svc_serv struct. In particular, ->sv_nrthreads but also to some - * extent ->sv_temp_socks and ->sv_permsocks. It also protects nfsdstats.th_cnt + * nfsd_mutex protects nn->nfsd_serv -- both the pointer itself and some members + * of the svc_serv struct such as ->sv_temp_socks and ->sv_permsocks. * * If (out side the lock) nn->nfsd_serv is non-NULL, then it must point to a - * properly initialised 'struct svc_serv' with ->sv_nrthreads > 0. That number - * of nfsd threads must exist and each must listed in ->sp_all_threads in each - * entry of ->sv_pools[]. + * properly initialised 'struct svc_serv' with ->sv_nrthreads > 0 (unless + * nn->keep_active is set). That number of nfsd threads must + * exist and each must be listed in ->sp_all_threads in some entry of + * ->sv_pools[]. * - * Transitions of the thread count between zero and non-zero are of particular - * interest since the svc_serv needs to be created and initialized at that - * point, or freed. + * Each active thread holds a counted reference on nn->nfsd_serv, as does + * the nn->keep_active flag and various transient calls to svc_get(). * * Finally, the nfsd_mutex also protects some of the global variables that are * accessed when nfsd starts and that are settable via the write_* routines in @@ -345,33 +345,57 @@ static bool nfsd_needs_lockd(struct nfsd_net *nn) return nfsd_vers(nn, 2, NFSD_TEST) || nfsd_vers(nn, 3, NFSD_TEST); } -void nfsd_copy_boot_verifier(__be32 verf[2], struct nfsd_net *nn) +/** + * nfsd_copy_write_verifier - Atomically copy a write verifier + * @verf: buffer in which to receive the verifier cookie + * @nn: NFS net namespace + * + * This function provides a wait-free mechanism for copying the + * namespace's write verifier without tearing it. + */ +void nfsd_copy_write_verifier(__be32 verf[2], struct nfsd_net *nn) { int seq = 0; do { - read_seqbegin_or_lock(&nn->boot_lock, &seq); - /* - * This is opaque to client, so no need to byte-swap. Use - * __force to keep sparse happy. y2038 time_t overflow is - * irrelevant in this usage - */ - verf[0] = (__force __be32)nn->nfssvc_boot.tv_sec; - verf[1] = (__force __be32)nn->nfssvc_boot.tv_nsec; - } while (need_seqretry(&nn->boot_lock, seq)); - done_seqretry(&nn->boot_lock, seq); + read_seqbegin_or_lock(&nn->writeverf_lock, &seq); + memcpy(verf, nn->writeverf, sizeof(*verf)); + } while (need_seqretry(&nn->writeverf_lock, seq)); + done_seqretry(&nn->writeverf_lock, seq); } -static void nfsd_reset_boot_verifier_locked(struct nfsd_net *nn) +static void nfsd_reset_write_verifier_locked(struct nfsd_net *nn) { - ktime_get_real_ts64(&nn->nfssvc_boot); + struct timespec64 now; + u64 verf; + + /* + * Because the time value is hashed, y2038 time_t overflow + * is irrelevant in this usage. + */ + ktime_get_raw_ts64(&now); + verf = siphash_2u64(now.tv_sec, now.tv_nsec, &nn->siphash_key); + memcpy(nn->writeverf, &verf, sizeof(nn->writeverf)); } -void nfsd_reset_boot_verifier(struct nfsd_net *nn) +/** + * nfsd_reset_write_verifier - Generate a new write verifier + * @nn: NFS net namespace + * + * This function updates the ->writeverf field of @nn. This field + * contains an opaque cookie that, according to Section 18.32.3 of + * RFC 8881, "the client can use to determine whether a server has + * changed instance state (e.g., server restart) between a call to + * WRITE and a subsequent call to either WRITE or COMMIT. This + * cookie MUST be unchanged during a single instance of the NFSv4.1 + * server and MUST be unique between instances of the NFSv4.1 + * server." + */ +void nfsd_reset_write_verifier(struct nfsd_net *nn) { - write_seqlock(&nn->boot_lock); - nfsd_reset_boot_verifier_locked(nn); - write_sequnlock(&nn->boot_lock); + write_seqlock(&nn->writeverf_lock); + nfsd_reset_write_verifier_locked(nn); + write_sequnlock(&nn->writeverf_lock); } static int nfsd_startup_net(struct net *net, const struct cred *cred) @@ -435,6 +459,7 @@ static void nfsd_shutdown_net(struct net *net) nfsd_shutdown_generic(); } +static DEFINE_SPINLOCK(nfsd_notifier_lock); static int nfsd_inetaddr_event(struct notifier_block *this, unsigned long event, void *ptr) { @@ -444,18 +469,17 @@ static int nfsd_inetaddr_event(struct notifier_block *this, unsigned long event, struct nfsd_net *nn = net_generic(net, nfsd_net_id); struct sockaddr_in sin; - if ((event != NETDEV_DOWN) || - !atomic_inc_not_zero(&nn->ntf_refcnt)) + if (event != NETDEV_DOWN || !nn->nfsd_serv) goto out; + spin_lock(&nfsd_notifier_lock); if (nn->nfsd_serv) { dprintk("nfsd_inetaddr_event: removed %pI4\n", &ifa->ifa_local); sin.sin_family = AF_INET; sin.sin_addr.s_addr = ifa->ifa_local; svc_age_temp_xprts_now(nn->nfsd_serv, (struct sockaddr *)&sin); } - atomic_dec(&nn->ntf_refcnt); - wake_up(&nn->ntf_wq); + spin_unlock(&nfsd_notifier_lock); out: return NOTIFY_DONE; @@ -475,10 +499,10 @@ static int nfsd_inet6addr_event(struct notifier_block *this, struct nfsd_net *nn = net_generic(net, nfsd_net_id); struct sockaddr_in6 sin6; - if ((event != NETDEV_DOWN) || - !atomic_inc_not_zero(&nn->ntf_refcnt)) + if (event != NETDEV_DOWN || !nn->nfsd_serv) goto out; + spin_lock(&nfsd_notifier_lock); if (nn->nfsd_serv) { dprintk("nfsd_inet6addr_event: removed %pI6\n", &ifa->addr); sin6.sin6_family = AF_INET6; @@ -487,8 +511,8 @@ static int nfsd_inet6addr_event(struct notifier_block *this, sin6.sin6_scope_id = ifa->idev->dev->ifindex; svc_age_temp_xprts_now(nn->nfsd_serv, (struct sockaddr *)&sin6); } - atomic_dec(&nn->ntf_refcnt); - wake_up(&nn->ntf_wq); + spin_unlock(&nfsd_notifier_lock); + out: return NOTIFY_DONE; } @@ -505,7 +529,6 @@ static void nfsd_last_thread(struct svc_serv *serv, struct net *net) { struct nfsd_net *nn = net_generic(net, nfsd_net_id); - atomic_dec(&nn->ntf_refcnt); /* check if the notifier still has clients */ if (atomic_dec_return(&nfsd_notifier_refcount) == 0) { unregister_inetaddr_notifier(&nfsd_inetaddr_notifier); @@ -513,7 +536,6 @@ static void nfsd_last_thread(struct svc_serv *serv, struct net *net) unregister_inet6addr_notifier(&nfsd_inet6addr_notifier); #endif } - wait_event(nn->ntf_wq, atomic_read(&nn->ntf_refcnt) == 0); /* * write_ports can create the server without actually starting @@ -594,20 +616,9 @@ static const struct svc_serv_ops nfsd_thread_sv_ops = { .svo_shutdown = nfsd_last_thread, .svo_function = nfsd, .svo_enqueue_xprt = svc_xprt_do_enqueue, - .svo_setup = svc_set_num_threads, .svo_module = THIS_MODULE, }; -static void nfsd_complete_shutdown(struct net *net) -{ - struct nfsd_net *nn = net_generic(net, nfsd_net_id); - - WARN_ON(!mutex_is_locked(&nfsd_mutex)); - - nn->nfsd_serv = NULL; - complete(&nn->nfsd_shutdown_complete); -} - void nfsd_shutdown_threads(struct net *net) { struct nfsd_net *nn = net_generic(net, nfsd_net_id); @@ -622,11 +633,9 @@ void nfsd_shutdown_threads(struct net *net) svc_get(serv); /* Kill outstanding nfsd threads */ - serv->sv_ops->svo_setup(serv, NULL, 0); - nfsd_destroy(net); + svc_set_num_threads(serv, NULL, 0); + nfsd_put(net); mutex_unlock(&nfsd_mutex); - /* Wait for shutdown of nfsd_serv to complete */ - wait_for_completion(&nn->nfsd_shutdown_complete); } bool i_am_nfsd(void) @@ -638,6 +647,7 @@ int nfsd_create_serv(struct net *net) { int error; struct nfsd_net *nn = net_generic(net, nfsd_net_id); + struct svc_serv *serv; WARN_ON(!mutex_is_locked(&nfsd_mutex)); if (nn->nfsd_serv) { @@ -647,19 +657,23 @@ int nfsd_create_serv(struct net *net) if (nfsd_max_blksize == 0) nfsd_max_blksize = nfsd_get_default_max_blksize(); nfsd_reset_versions(nn); - nn->nfsd_serv = svc_create_pooled(&nfsd_program, nfsd_max_blksize, - &nfsd_thread_sv_ops); - if (nn->nfsd_serv == NULL) + serv = svc_create_pooled(&nfsd_program, nfsd_max_blksize, + &nfsd_thread_sv_ops); + if (serv == NULL) return -ENOMEM; - init_completion(&nn->nfsd_shutdown_complete); - nn->nfsd_serv->sv_maxconn = nn->max_connections; - error = svc_bind(nn->nfsd_serv, net); + serv->sv_maxconn = nn->max_connections; + error = svc_bind(serv, net); if (error < 0) { - svc_destroy(nn->nfsd_serv); - nfsd_complete_shutdown(net); + /* NOT nfsd_put() as notifiers (see below) haven't + * been set up yet. + */ + svc_put(serv); return error; } + spin_lock(&nfsd_notifier_lock); + nn->nfsd_serv = serv; + spin_unlock(&nfsd_notifier_lock); set_max_drc(); /* check if the notifier is already set */ @@ -669,8 +683,7 @@ int nfsd_create_serv(struct net *net) register_inet6addr_notifier(&nfsd_inet6addr_notifier); #endif } - atomic_inc(&nn->ntf_refcnt); - nfsd_reset_boot_verifier(nn); + nfsd_reset_write_verifier(nn); return 0; } @@ -697,16 +710,26 @@ int nfsd_get_nrthreads(int n, int *nthreads, struct net *net) return 0; } -void nfsd_destroy(struct net *net) +/* This is the callback for kref_put() below. + * There is no code here as the first thing to be done is + * call svc_shutdown_net(), but we cannot get the 'net' from + * the kref. So do all the work when kref_put returns true. + */ +static void nfsd_noop(struct kref *ref) +{ +} + +void nfsd_put(struct net *net) { struct nfsd_net *nn = net_generic(net, nfsd_net_id); - int destroy = (nn->nfsd_serv->sv_nrthreads == 1); - if (destroy) + if (kref_put(&nn->nfsd_serv->sv_refcnt, nfsd_noop)) { svc_shutdown_net(nn->nfsd_serv, net); - svc_destroy(nn->nfsd_serv); - if (destroy) - nfsd_complete_shutdown(net); + svc_destroy(&nn->nfsd_serv->sv_refcnt); + spin_lock(&nfsd_notifier_lock); + nn->nfsd_serv = NULL; + spin_unlock(&nfsd_notifier_lock); + } } int nfsd_set_nrthreads(int n, int *nthreads, struct net *net) @@ -733,7 +756,7 @@ int nfsd_set_nrthreads(int n, int *nthreads, struct net *net) if (tot > NFSD_MAXSERVS) { /* total too large: scale down requested numbers */ for (i = 0; i < n && tot > 0; i++) { - int new = nthreads[i] * NFSD_MAXSERVS / tot; + int new = nthreads[i] * NFSD_MAXSERVS / tot; tot -= (nthreads[i] - new); nthreads[i] = new; } @@ -753,12 +776,13 @@ int nfsd_set_nrthreads(int n, int *nthreads, struct net *net) /* apply the new numbers */ svc_get(nn->nfsd_serv); for (i = 0; i < n; i++) { - err = nn->nfsd_serv->sv_ops->svo_setup(nn->nfsd_serv, - &nn->nfsd_serv->sv_pools[i], nthreads[i]); + err = svc_set_num_threads(nn->nfsd_serv, + &nn->nfsd_serv->sv_pools[i], + nthreads[i]); if (err) break; } - nfsd_destroy(net); + nfsd_put(net); return err; } @@ -795,21 +819,19 @@ nfsd_svc(int nrservs, struct net *net, const struct cred *cred) error = nfsd_startup_net(net, cred); if (error) - goto out_destroy; - error = nn->nfsd_serv->sv_ops->svo_setup(nn->nfsd_serv, - NULL, nrservs); + goto out_put; + error = svc_set_num_threads(nn->nfsd_serv, NULL, nrservs); if (error) goto out_shutdown; - /* We are holding a reference to nn->nfsd_serv which - * we don't want to count in the return value, - * so subtract 1 - */ - error = nn->nfsd_serv->sv_nrthreads - 1; + error = nn->nfsd_serv->sv_nrthreads; out_shutdown: if (error < 0 && !nfsd_up_before) nfsd_shutdown_net(net); -out_destroy: - nfsd_destroy(net); /* Release server */ +out_put: + /* Threads now hold service active */ + if (xchg(&nn->keep_active, 0)) + nfsd_put(net); + nfsd_put(net); out: mutex_unlock(&nfsd_mutex); return error; @@ -923,9 +945,6 @@ nfsd(void *vrqstp) struct nfsd_net *nn = net_generic(net, nfsd_net_id); int err; - /* Lock module and set up kernel thread */ - mutex_lock(&nfsd_mutex); - /* At this point, the thread shares current->fs * with the init process. We need to create files with the * umask as defined by the client instead of init's umask. */ @@ -945,8 +964,7 @@ nfsd(void *vrqstp) allow_signal(SIGINT); allow_signal(SIGQUIT); - nfsdstats.th_cnt++; - mutex_unlock(&nfsd_mutex); + atomic_inc(&nfsdstats.th_cnt); set_freezable(); @@ -973,20 +991,36 @@ nfsd(void *vrqstp) /* Clear signals before calling svc_exit_thread() */ flush_signals(current); - mutex_lock(&nfsd_mutex); - nfsdstats.th_cnt --; + atomic_dec(&nfsdstats.th_cnt); out: - rqstp->rq_server = NULL; + /* Take an extra ref so that the svc_put in svc_exit_thread() + * doesn't call svc_destroy() + */ + svc_get(nn->nfsd_serv); /* Release the thread */ svc_exit_thread(rqstp); - nfsd_destroy(net); + /* We need to drop a ref, but may not drop the last reference + * without holding nfsd_mutex, and we cannot wait for nfsd_mutex as that + * could deadlock with nfsd_shutdown_threads() waiting for us. + * So three options are: + * - drop a non-final reference, + * - get the mutex without waiting + * - sleep briefly andd try the above again + */ + while (!svc_put_not_last(nn->nfsd_serv)) { + if (mutex_trylock(&nfsd_mutex)) { + nfsd_put(net); + mutex_unlock(&nfsd_mutex); + break; + } + msleep(20); + } /* Release module */ - mutex_unlock(&nfsd_mutex); - module_put_and_exit(0); + module_put_and_kthread_exit(0); return 0; } @@ -1096,7 +1130,6 @@ int nfsd_pool_stats_open(struct inode *inode, struct file *file) mutex_unlock(&nfsd_mutex); return -ENODEV; } - /* bump up the psudo refcount while traversing */ svc_get(nn->nfsd_serv); ret = svc_pool_stats_open(nn->nfsd_serv, file); mutex_unlock(&nfsd_mutex); @@ -1109,8 +1142,7 @@ int nfsd_pool_stats_release(struct inode *inode, struct file *file) struct net *net = inode->i_sb->s_fs_info; mutex_lock(&nfsd_mutex); - /* this function really, really should have been called svc_put() */ - nfsd_destroy(net); + nfsd_put(net); mutex_unlock(&nfsd_mutex); return ret; } diff --git a/fs/nfsd/state.h b/fs/nfsd/state.h index e73bdbb1634a..95457cfd37fc 100644 --- a/fs/nfsd/state.h +++ b/fs/nfsd/state.h @@ -568,6 +568,10 @@ struct nfs4_ol_stateid { struct list_head st_locks; struct nfs4_stateowner *st_stateowner; struct nfs4_clnt_odstate *st_clnt_odstate; +/* + * These bitmasks use 3 separate bits for READ, ALLOW, and BOTH; see the + * comment above bmap_to_share_mode() for explanation: + */ unsigned char st_access_bmap; unsigned char st_deny_bmap; struct nfs4_ol_stateid *st_openstp; @@ -629,6 +633,7 @@ struct nfsd4_blocked_lock { struct file_lock nbl_lock; struct knfsd_fh nbl_fh; struct nfsd4_callback nbl_cb; + struct kref nbl_kref; }; struct nfsd4_compound_state; diff --git a/fs/nfsd/stats.c b/fs/nfsd/stats.c index 1d3b881e7382..a8c5a02a84f0 100644 --- a/fs/nfsd/stats.c +++ b/fs/nfsd/stats.c @@ -45,7 +45,7 @@ static int nfsd_proc_show(struct seq_file *seq, void *v) percpu_counter_sum_positive(&nfsdstats.counter[NFSD_STATS_IO_WRITE])); /* thread usage: */ - seq_printf(seq, "th %u 0", nfsdstats.th_cnt); + seq_printf(seq, "th %u 0", atomic_read(&nfsdstats.th_cnt)); /* deprecated thread usage histogram stats */ for (i = 0; i < 10; i++) diff --git a/fs/nfsd/stats.h b/fs/nfsd/stats.h index 51ecda852e23..9b43dc3d9991 100644 --- a/fs/nfsd/stats.h +++ b/fs/nfsd/stats.h @@ -29,11 +29,9 @@ enum { struct nfsd_stats { struct percpu_counter counter[NFSD_STATS_COUNTERS_NUM]; - /* Protected by nfsd_mutex */ - unsigned int th_cnt; /* number of available threads */ + atomic_t th_cnt; /* number of available threads */ }; - extern struct nfsd_stats nfsdstats; extern struct svc_stat nfsd_svcstats; diff --git a/fs/nfsd/trace.h b/fs/nfsd/trace.h index f1e0d3c51bc2..c4cf56327843 100644 --- a/fs/nfsd/trace.h +++ b/fs/nfsd/trace.h @@ -47,7 +47,7 @@ rqstp->rq_xprt->xpt_remotelen); \ } while (0); -TRACE_EVENT(nfsd_garbage_args_err, +DECLARE_EVENT_CLASS(nfsd_xdr_err_class, TP_PROTO( const struct svc_rqst *rqstp ), @@ -69,27 +69,13 @@ TRACE_EVENT(nfsd_garbage_args_err, ) ); -TRACE_EVENT(nfsd_cant_encode_err, - TP_PROTO( - const struct svc_rqst *rqstp - ), - TP_ARGS(rqstp), - TP_STRUCT__entry( - NFSD_TRACE_PROC_ARG_FIELDS - - __field(u32, vers) - __field(u32, proc) - ), - TP_fast_assign( - NFSD_TRACE_PROC_ARG_ASSIGNMENTS +#define DEFINE_NFSD_XDR_ERR_EVENT(name) \ +DEFINE_EVENT(nfsd_xdr_err_class, nfsd_##name##_err, \ + TP_PROTO(const struct svc_rqst *rqstp), \ + TP_ARGS(rqstp)) - __entry->vers = rqstp->rq_vers; - __entry->proc = rqstp->rq_proc; - ), - TP_printk("xid=0x%08x vers=%u proc=%u", - __entry->xid, __entry->vers, __entry->proc - ) -); +DEFINE_NFSD_XDR_ERR_EVENT(garbage_args); +DEFINE_NFSD_XDR_ERR_EVENT(cant_encode); #define show_nfsd_may_flags(x) \ __print_flags(x, "|", \ @@ -413,6 +399,56 @@ TRACE_EVENT(nfsd_dirent, ) ) +DECLARE_EVENT_CLASS(nfsd_copy_err_class, + TP_PROTO(struct svc_rqst *rqstp, + struct svc_fh *src_fhp, + loff_t src_offset, + struct svc_fh *dst_fhp, + loff_t dst_offset, + u64 count, + int status), + TP_ARGS(rqstp, src_fhp, src_offset, dst_fhp, dst_offset, count, status), + TP_STRUCT__entry( + __field(u32, xid) + __field(u32, src_fh_hash) + __field(loff_t, src_offset) + __field(u32, dst_fh_hash) + __field(loff_t, dst_offset) + __field(u64, count) + __field(int, status) + ), + TP_fast_assign( + __entry->xid = be32_to_cpu(rqstp->rq_xid); + __entry->src_fh_hash = knfsd_fh_hash(&src_fhp->fh_handle); + __entry->src_offset = src_offset; + __entry->dst_fh_hash = knfsd_fh_hash(&dst_fhp->fh_handle); + __entry->dst_offset = dst_offset; + __entry->count = count; + __entry->status = status; + ), + TP_printk("xid=0x%08x src_fh_hash=0x%08x src_offset=%lld " + "dst_fh_hash=0x%08x dst_offset=%lld " + "count=%llu status=%d", + __entry->xid, __entry->src_fh_hash, __entry->src_offset, + __entry->dst_fh_hash, __entry->dst_offset, + (unsigned long long)__entry->count, + __entry->status) +) + +#define DEFINE_NFSD_COPY_ERR_EVENT(name) \ +DEFINE_EVENT(nfsd_copy_err_class, nfsd_##name, \ + TP_PROTO(struct svc_rqst *rqstp, \ + struct svc_fh *src_fhp, \ + loff_t src_offset, \ + struct svc_fh *dst_fhp, \ + loff_t dst_offset, \ + u64 count, \ + int status), \ + TP_ARGS(rqstp, src_fhp, src_offset, dst_fhp, dst_offset, \ + count, status)) + +DEFINE_NFSD_COPY_ERR_EVENT(clone_file_range_err); + #include "state.h" #include "filecache.h" #include "vfs.h" @@ -538,6 +574,34 @@ DEFINE_EVENT(nfsd_net_class, nfsd_##name, \ DEFINE_NET_EVENT(grace_start); DEFINE_NET_EVENT(grace_complete); +TRACE_EVENT(nfsd_writeverf_reset, + TP_PROTO( + const struct nfsd_net *nn, + const struct svc_rqst *rqstp, + int error + ), + TP_ARGS(nn, rqstp, error), + TP_STRUCT__entry( + __field(unsigned long long, boot_time) + __field(u32, xid) + __field(int, error) + __array(unsigned char, verifier, NFS4_VERIFIER_SIZE) + ), + TP_fast_assign( + __entry->boot_time = nn->boot_time; + __entry->xid = be32_to_cpu(rqstp->rq_xid); + __entry->error = error; + + /* avoid seqlock inside TP_fast_assign */ + memcpy(__entry->verifier, nn->writeverf, + NFS4_VERIFIER_SIZE); + ), + TP_printk("boot_time=%16llx xid=0x%08x error=%d new verifier=0x%s", + __entry->boot_time, __entry->xid, __entry->error, + __print_hex_str(__entry->verifier, NFS4_VERIFIER_SIZE) + ) +); + TRACE_EVENT(nfsd_clid_cred_mismatch, TP_PROTO( const struct nfs4_client *clp, diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index c99857689e2c..99c2b9dfbb10 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -40,6 +40,7 @@ #include "../internal.h" #include "acl.h" #include "idmap.h" +#include "xdr4.h" #endif /* CONFIG_NFSD_V4 */ #include "nfsd.h" @@ -517,15 +518,23 @@ __be32 nfsd4_set_nfs4_label(struct svc_rqst *rqstp, struct svc_fh *fhp, } #endif -__be32 nfsd4_clone_file_range(struct nfsd_file *nf_src, u64 src_pos, - struct nfsd_file *nf_dst, u64 dst_pos, u64 count, bool sync) +static struct nfsd4_compound_state *nfsd4_get_cstate(struct svc_rqst *rqstp) +{ + return &((struct nfsd4_compoundres *)rqstp->rq_resp)->cstate; +} + +__be32 nfsd4_clone_file_range(struct svc_rqst *rqstp, + struct nfsd_file *nf_src, u64 src_pos, + struct nfsd_file *nf_dst, u64 dst_pos, + u64 count, bool sync) { struct file *src = nf_src->nf_file; struct file *dst = nf_dst->nf_file; + errseq_t since; loff_t cloned; __be32 ret = 0; - down_write(&nf_dst->nf_rwsem); + since = READ_ONCE(dst->f_wb_err); cloned = vfs_clone_file_range(src, src_pos, dst, dst_pos, count, 0); if (cloned < 0) { ret = nfserrno(cloned); @@ -540,15 +549,25 @@ __be32 nfsd4_clone_file_range(struct nfsd_file *nf_src, u64 src_pos, int status = vfs_fsync_range(dst, dst_pos, dst_end, 0); if (!status) + status = filemap_check_wb_err(dst->f_mapping, since); + if (!status) status = commit_inode_metadata(file_inode(src)); if (status < 0) { - nfsd_reset_boot_verifier(net_generic(nf_dst->nf_net, - nfsd_net_id)); + struct nfsd_net *nn = net_generic(nf_dst->nf_net, + nfsd_net_id); + + trace_nfsd_clone_file_range_err(rqstp, + &nfsd4_get_cstate(rqstp)->save_fh, + src_pos, + &nfsd4_get_cstate(rqstp)->current_fh, + dst_pos, + count, status); + nfsd_reset_write_verifier(nn); + trace_nfsd_writeverf_reset(nn, rqstp, status); ret = nfserrno(status); } } out_err: - up_write(&nf_dst->nf_rwsem); return ret; } @@ -777,6 +796,7 @@ nfsd_open(struct svc_rqst *rqstp, struct svc_fh *fhp, umode_t type, int may_flags, struct file **filp) { __be32 err; + bool retried = false; validate_process_creds(); /* @@ -792,9 +812,16 @@ nfsd_open(struct svc_rqst *rqstp, struct svc_fh *fhp, umode_t type, */ if (type == S_IFREG) may_flags |= NFSD_MAY_OWNER_OVERRIDE; +retry: err = fh_verify(rqstp, fhp, type, may_flags); - if (!err) + if (!err) { err = __nfsd_open(rqstp, fhp, type, may_flags, filp); + if (err == nfserr_stale && !retried) { + retried = true; + fh_put(fhp); + goto retry; + } + } validate_process_creds(); return err; } @@ -944,10 +971,12 @@ nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct nfsd_file *nf, unsigned long *cnt, int stable, __be32 *verf) { + struct nfsd_net *nn = net_generic(SVC_NET(rqstp), nfsd_net_id); struct file *file = nf->nf_file; struct super_block *sb = file_inode(file)->i_sb; struct svc_export *exp; struct iov_iter iter; + errseq_t since; __be32 nfserr; int host_err; int use_wgather; @@ -985,36 +1014,28 @@ nfsd_vfs_write(struct svc_rqst *rqstp, struct svc_fh *fhp, struct nfsd_file *nf, flags |= RWF_SYNC; iov_iter_kvec(&iter, WRITE, vec, vlen, *cnt); - if (flags & RWF_SYNC) { - down_write(&nf->nf_rwsem); - host_err = vfs_iter_write(file, &iter, &pos, flags); - if (host_err < 0) - nfsd_reset_boot_verifier(net_generic(SVC_NET(rqstp), - nfsd_net_id)); - up_write(&nf->nf_rwsem); - } else { - down_read(&nf->nf_rwsem); - if (verf) - nfsd_copy_boot_verifier(verf, - net_generic(SVC_NET(rqstp), - nfsd_net_id)); - host_err = vfs_iter_write(file, &iter, &pos, flags); - up_read(&nf->nf_rwsem); - } + since = READ_ONCE(file->f_wb_err); + if (verf) + nfsd_copy_write_verifier(verf, nn); + host_err = vfs_iter_write(file, &iter, &pos, flags); if (host_err < 0) { - nfsd_reset_boot_verifier(net_generic(SVC_NET(rqstp), - nfsd_net_id)); + nfsd_reset_write_verifier(nn); + trace_nfsd_writeverf_reset(nn, rqstp, host_err); goto out_nfserr; } *cnt = host_err; nfsd_stats_io_write_add(exp, *cnt); fsnotify_modify(file); + host_err = filemap_check_wb_err(file->f_mapping, since); + if (host_err < 0) + goto out_nfserr; if (stable && use_wgather) { host_err = wait_for_concurrent_writes(file); - if (host_err < 0) - nfsd_reset_boot_verifier(net_generic(SVC_NET(rqstp), - nfsd_net_id)); + if (host_err < 0) { + nfsd_reset_write_verifier(nn); + trace_nfsd_writeverf_reset(nn, rqstp, host_err); + } } out_nfserr: @@ -1089,19 +1110,6 @@ out: } #ifdef CONFIG_NFSD_V3 -static int -nfsd_filemap_write_and_wait_range(struct nfsd_file *nf, loff_t offset, - loff_t end) -{ - struct address_space *mapping = nf->nf_file->f_mapping; - int ret = filemap_fdatawrite_range(mapping, offset, end); - - if (ret) - return ret; - filemap_fdatawait_range_keep_errors(mapping, offset, end); - return 0; -} - /* * Commit all pending writes to stable storage. * @@ -1115,6 +1123,7 @@ __be32 nfsd_commit(struct svc_rqst *rqstp, struct svc_fh *fhp, loff_t offset, unsigned long count, __be32 *verf) { + struct nfsd_net *nn; struct nfsd_file *nf; loff_t end = LLONG_MAX; __be32 err = nfserr_inval; @@ -1131,29 +1140,28 @@ nfsd_commit(struct svc_rqst *rqstp, struct svc_fh *fhp, NFSD_MAY_WRITE|NFSD_MAY_NOT_BREAK_LEASE, &nf); if (err) goto out; + nn = net_generic(nf->nf_net, nfsd_net_id); if (EX_ISSYNC(fhp->fh_export)) { - int err2 = nfsd_filemap_write_and_wait_range(nf, offset, end); + errseq_t since = READ_ONCE(nf->nf_file->f_wb_err); + int err2; - down_write(&nf->nf_rwsem); - if (!err2) - err2 = vfs_fsync_range(nf->nf_file, offset, end, 0); + err2 = vfs_fsync_range(nf->nf_file, offset, end, 0); switch (err2) { case 0: - nfsd_copy_boot_verifier(verf, net_generic(nf->nf_net, - nfsd_net_id)); + nfsd_copy_write_verifier(verf, nn); + err2 = filemap_check_wb_err(nf->nf_file->f_mapping, + since); break; case -EINVAL: err = nfserr_notsupp; break; default: - err = nfserrno(err2); - nfsd_reset_boot_verifier(net_generic(nf->nf_net, - nfsd_net_id)); + nfsd_reset_write_verifier(nn); + trace_nfsd_writeverf_reset(nn, rqstp, err2); } - up_write(&nf->nf_rwsem); + err = nfserrno(err2); } else - nfsd_copy_boot_verifier(verf, net_generic(nf->nf_net, - nfsd_net_id)); + nfsd_copy_write_verifier(verf, nn); nfsd_file_put(nf); out: @@ -1747,8 +1755,8 @@ retry: * so do it by hand */ trap = lock_rename(tdentry, fdentry); ffhp->fh_locked = tfhp->fh_locked = true; - fill_pre_wcc(ffhp); - fill_pre_wcc(tfhp); + fh_fill_pre_attrs(ffhp); + fh_fill_pre_attrs(tfhp); odentry = lookup_one_len(fname, fdentry, flen); host_err = PTR_ERR(odentry); @@ -1808,8 +1816,8 @@ retry: * were the same, so again we do it by hand. */ if (!close_cached) { - fill_post_wcc(ffhp); - fill_post_wcc(tfhp); + fh_fill_post_attrs(ffhp); + fh_fill_post_attrs(tfhp); } unlock_rename(tdentry, fdentry); ffhp->fh_locked = tfhp->fh_locked = false; diff --git a/fs/nfsd/vfs.h b/fs/nfsd/vfs.h index b21b76e6b9a8..9f56dcb22ff7 100644 --- a/fs/nfsd/vfs.h +++ b/fs/nfsd/vfs.h @@ -57,7 +57,8 @@ __be32 nfsd4_set_nfs4_label(struct svc_rqst *, struct svc_fh *, struct xdr_netobj *); __be32 nfsd4_vfs_fallocate(struct svc_rqst *, struct svc_fh *, struct file *, loff_t, loff_t, int); -__be32 nfsd4_clone_file_range(struct nfsd_file *nf_src, u64 src_pos, +__be32 nfsd4_clone_file_range(struct svc_rqst *rqstp, + struct nfsd_file *nf_src, u64 src_pos, struct nfsd_file *nf_dst, u64 dst_pos, u64 count, bool sync); #endif /* CONFIG_NFSD_V4 */ diff --git a/fs/nilfs2/page.c b/fs/nilfs2/page.c index bc3e2cd4117f..063dd16d75b5 100644 --- a/fs/nilfs2/page.c +++ b/fs/nilfs2/page.c @@ -195,12 +195,12 @@ void nilfs_page_bug(struct page *page) */ static void nilfs_copy_page(struct page *dst, struct page *src, int copy_dirty) { - struct buffer_head *dbh, *dbufs, *sbh, *sbufs; + struct buffer_head *dbh, *dbufs, *sbh; unsigned long mask = NILFS_BUFFER_INHERENT_BITS; BUG_ON(PageWriteback(dst)); - sbh = sbufs = page_buffers(src); + sbh = page_buffers(src); if (!page_has_buffers(dst)) create_empty_buffers(dst, sbh->b_size, 0); diff --git a/fs/notify/dnotify/dnotify.c b/fs/notify/dnotify/dnotify.c index d5ebebb034ff..829dd4a61b66 100644 --- a/fs/notify/dnotify/dnotify.c +++ b/fs/notify/dnotify/dnotify.c @@ -19,7 +19,25 @@ #include <linux/fdtable.h> #include <linux/fsnotify_backend.h> -int dir_notify_enable __read_mostly = 1; +static int dir_notify_enable __read_mostly = 1; +#ifdef CONFIG_SYSCTL +static struct ctl_table dnotify_sysctls[] = { + { + .procname = "dir-notify-enable", + .data = &dir_notify_enable, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec, + }, + {} +}; +static void __init dnotify_sysctl_init(void) +{ + register_sysctl_init("fs", dnotify_sysctls); +} +#else +#define dnotify_sysctl_init() do { } while (0) +#endif static struct kmem_cache *dnotify_struct_cache __read_mostly; static struct kmem_cache *dnotify_mark_cache __read_mostly; @@ -386,6 +404,7 @@ static int __init dnotify_init(void) dnotify_group = fsnotify_alloc_group(&dnotify_fsnotify_ops); if (IS_ERR(dnotify_group)) panic("unable to allocate fsnotify group for dnotify\n"); + dnotify_sysctl_init(); return 0; } diff --git a/fs/notify/fanotify/fanotify_user.c b/fs/notify/fanotify/fanotify_user.c index 73a3e939c921..73b1615f9d96 100644 --- a/fs/notify/fanotify/fanotify_user.c +++ b/fs/notify/fanotify/fanotify_user.c @@ -59,7 +59,7 @@ static int fanotify_max_queued_events __read_mostly; static long ft_zero = 0; static long ft_int_max = INT_MAX; -struct ctl_table fanotify_table[] = { +static struct ctl_table fanotify_table[] = { { .procname = "max_user_groups", .data = &init_user_ns.ucount_max[UCOUNT_FANOTIFY_GROUPS], @@ -88,6 +88,13 @@ struct ctl_table fanotify_table[] = { }, { } }; + +static void __init fanotify_sysctls_init(void) +{ + register_sysctl("fs/fanotify", fanotify_table); +} +#else +#define fanotify_sysctls_init() do { } while (0) #endif /* CONFIG_SYSCTL */ /* @@ -1743,6 +1750,7 @@ static int __init fanotify_user_setup(void) init_user_ns.ucount_max[UCOUNT_FANOTIFY_GROUPS] = FANOTIFY_DEFAULT_MAX_GROUPS; init_user_ns.ucount_max[UCOUNT_FANOTIFY_MARKS] = max_marks; + fanotify_sysctls_init(); return 0; } diff --git a/fs/notify/inotify/inotify_user.c b/fs/notify/inotify/inotify_user.c index 29fca3284bb5..54583f62dc44 100644 --- a/fs/notify/inotify/inotify_user.c +++ b/fs/notify/inotify/inotify_user.c @@ -58,7 +58,7 @@ struct kmem_cache *inotify_inode_mark_cachep __read_mostly; static long it_zero = 0; static long it_int_max = INT_MAX; -struct ctl_table inotify_table[] = { +static struct ctl_table inotify_table[] = { { .procname = "max_user_instances", .data = &init_user_ns.ucount_max[UCOUNT_INOTIFY_INSTANCES], @@ -87,6 +87,14 @@ struct ctl_table inotify_table[] = { }, { } }; + +static void __init inotify_sysctls_init(void) +{ + register_sysctl("fs/inotify", inotify_table); +} + +#else +#define inotify_sysctls_init() do { } while (0) #endif /* CONFIG_SYSCTL */ static inline __u32 inotify_arg_to_mask(struct inode *inode, u32 arg) @@ -849,6 +857,7 @@ static int __init inotify_user_setup(void) inotify_max_queued_events = 16384; init_user_ns.ucount_max[UCOUNT_INOTIFY_INSTANCES] = 128; init_user_ns.ucount_max[UCOUNT_INOTIFY_WATCHES] = watches_max; + inotify_sysctls_init(); return 0; } diff --git a/fs/ntfs/attrib.c b/fs/ntfs/attrib.c index d563abc3e136..2911c04a33e0 100644 --- a/fs/ntfs/attrib.c +++ b/fs/ntfs/attrib.c @@ -1,5 +1,5 @@ // SPDX-License-Identifier: GPL-2.0-or-later -/** +/* * attrib.c - NTFS attribute operations. Part of the Linux-NTFS project. * * Copyright (c) 2001-2012 Anton Altaparmakov and Tuxera Inc. diff --git a/fs/ntfs3/ntfs_fs.h b/fs/ntfs3/ntfs_fs.h index 8aaec7e0804e..fb825059d488 100644 --- a/fs/ntfs3/ntfs_fs.h +++ b/fs/ntfs3/ntfs_fs.h @@ -11,7 +11,6 @@ #include <linux/blkdev.h> #include <linux/buffer_head.h> -#include <linux/cleancache.h> #include <linux/fs.h> #include <linux/highmem.h> #include <linux/kernel.h> diff --git a/fs/ocfs2/alloc.c b/fs/ocfs2/alloc.c index bb247bc349e4..bf9357123bc5 100644 --- a/fs/ocfs2/alloc.c +++ b/fs/ocfs2/alloc.c @@ -2040,7 +2040,7 @@ static void ocfs2_complete_edge_insert(handle_t *handle, int i, idx; struct ocfs2_extent_list *el, *left_el, *right_el; struct ocfs2_extent_rec *left_rec, *right_rec; - struct buffer_head *root_bh = left_path->p_node[subtree_index].bh; + struct buffer_head *root_bh; /* * Update the counts and position values within all the diff --git a/fs/ocfs2/aops.c b/fs/ocfs2/aops.c index 68d11c295dd3..498da317580a 100644 --- a/fs/ocfs2/aops.c +++ b/fs/ocfs2/aops.c @@ -1799,20 +1799,20 @@ try_again: */ ret = ocfs2_grab_pages_for_write(mapping, wc, wc->w_cpos, pos, len, cluster_of_pages, mmap_page); - if (ret && ret != -EAGAIN) { - mlog_errno(ret); - goto out_quota; - } + if (ret) { + /* + * ocfs2_grab_pages_for_write() returns -EAGAIN if it could not lock + * the target page. In this case, we exit with no error and no target + * page. This will trigger the caller, page_mkwrite(), to re-try + * the operation. + */ + if (type == OCFS2_WRITE_MMAP && ret == -EAGAIN) { + BUG_ON(wc->w_target_page); + ret = 0; + goto out_quota; + } - /* - * ocfs2_grab_pages_for_write() returns -EAGAIN if it could not lock - * the target page. In this case, we exit with no error and no target - * page. This will trigger the caller, page_mkwrite(), to re-try - * the operation. - */ - if (ret == -EAGAIN) { - BUG_ON(wc->w_target_page); - ret = 0; + mlog_errno(ret); goto out_quota; } diff --git a/fs/ocfs2/cluster/heartbeat.c b/fs/ocfs2/cluster/heartbeat.c index f89ffcbd585f..a17be1618bf7 100644 --- a/fs/ocfs2/cluster/heartbeat.c +++ b/fs/ocfs2/cluster/heartbeat.c @@ -379,7 +379,7 @@ static void o2hb_nego_timeout(struct work_struct *work) o2hb_fill_node_map(live_node_bitmap, sizeof(live_node_bitmap)); /* lowest node as master node to make negotiate decision. */ - master_node = find_next_bit(live_node_bitmap, O2NM_MAX_NODES, 0); + master_node = find_first_bit(live_node_bitmap, O2NM_MAX_NODES); if (master_node == o2nm_this_node()) { if (!test_bit(master_node, reg->hr_nego_node_bitmap)) { diff --git a/fs/ocfs2/cluster/masklog.c b/fs/ocfs2/cluster/masklog.c index 810d32815593..563881ddbf00 100644 --- a/fs/ocfs2/cluster/masklog.c +++ b/fs/ocfs2/cluster/masklog.c @@ -120,7 +120,8 @@ static struct mlog_attribute mlog_attrs[MLOG_MAX_BITS] = { define_mask(KTHREAD), }; -static struct attribute *mlog_attr_ptrs[MLOG_MAX_BITS] = {NULL, }; +static struct attribute *mlog_default_attrs[MLOG_MAX_BITS] = {NULL, }; +ATTRIBUTE_GROUPS(mlog_default); static ssize_t mlog_show(struct kobject *obj, struct attribute *attr, char *buf) @@ -144,8 +145,8 @@ static const struct sysfs_ops mlog_attr_ops = { }; static struct kobj_type mlog_ktype = { - .default_attrs = mlog_attr_ptrs, - .sysfs_ops = &mlog_attr_ops, + .default_groups = mlog_default_groups, + .sysfs_ops = &mlog_attr_ops, }; static struct kset mlog_kset = { @@ -157,10 +158,10 @@ int mlog_sys_init(struct kset *o2cb_kset) int i = 0; while (mlog_attrs[i].attr.mode) { - mlog_attr_ptrs[i] = &mlog_attrs[i].attr; + mlog_default_attrs[i] = &mlog_attrs[i].attr; i++; } - mlog_attr_ptrs[i] = NULL; + mlog_default_attrs[i] = NULL; kobject_set_name(&mlog_kset.kobj, "logmask"); mlog_kset.kobj.kset = o2cb_kset; diff --git a/fs/ocfs2/dir.c b/fs/ocfs2/dir.c index bd8d534f11cb..f2cc1ff29e6d 100644 --- a/fs/ocfs2/dir.c +++ b/fs/ocfs2/dir.c @@ -3343,7 +3343,7 @@ static int ocfs2_find_dir_space_id(struct inode *dir, struct buffer_head *di_bh, struct ocfs2_dir_entry *de, *last_de = NULL; char *de_buf, *limit; unsigned long offset = 0; - unsigned int rec_len, new_rec_len, free_space = dir->i_sb->s_blocksize; + unsigned int rec_len, new_rec_len, free_space; /* * This calculates how many free bytes we'd have in block zero, should diff --git a/fs/ocfs2/dlm/dlmdomain.c b/fs/ocfs2/dlm/dlmdomain.c index 9f90fc9551e1..c4eccd499db8 100644 --- a/fs/ocfs2/dlm/dlmdomain.c +++ b/fs/ocfs2/dlm/dlmdomain.c @@ -1045,7 +1045,7 @@ static int dlm_send_regions(struct dlm_ctxt *dlm, unsigned long *node_map) int status, ret = 0, i; char *p; - if (find_next_bit(node_map, O2NM_MAX_NODES, 0) >= O2NM_MAX_NODES) + if (find_first_bit(node_map, O2NM_MAX_NODES) >= O2NM_MAX_NODES) goto bail; qr = kzalloc(sizeof(struct dlm_query_region), GFP_KERNEL); @@ -1217,7 +1217,7 @@ static int dlm_send_nodeinfo(struct dlm_ctxt *dlm, unsigned long *node_map) struct o2nm_node *node; int ret = 0, status, count, i; - if (find_next_bit(node_map, O2NM_MAX_NODES, 0) >= O2NM_MAX_NODES) + if (find_first_bit(node_map, O2NM_MAX_NODES) >= O2NM_MAX_NODES) goto bail; qn = kzalloc(sizeof(struct dlm_query_nodeinfo), GFP_KERNEL); diff --git a/fs/ocfs2/dlm/dlmmaster.c b/fs/ocfs2/dlm/dlmmaster.c index 9b88219febb5..227da5b1b6ab 100644 --- a/fs/ocfs2/dlm/dlmmaster.c +++ b/fs/ocfs2/dlm/dlmmaster.c @@ -861,7 +861,7 @@ lookup: * to see if there are any nodes that still need to be * considered. these will not appear in the mle nodemap * but they might own this lockres. wait on them. */ - bit = find_next_bit(dlm->recovery_map, O2NM_MAX_NODES, 0); + bit = find_first_bit(dlm->recovery_map, O2NM_MAX_NODES); if (bit < O2NM_MAX_NODES) { mlog(0, "%s: res %.*s, At least one node (%d) " "to recover before lock mastery can begin\n", @@ -912,7 +912,7 @@ redo_request: dlm_wait_for_recovery(dlm); spin_lock(&dlm->spinlock); - bit = find_next_bit(dlm->recovery_map, O2NM_MAX_NODES, 0); + bit = find_first_bit(dlm->recovery_map, O2NM_MAX_NODES); if (bit < O2NM_MAX_NODES) { mlog(0, "%s: res %.*s, At least one node (%d) " "to recover before lock mastery can begin\n", @@ -1079,7 +1079,7 @@ recheck: sleep = 1; /* have all nodes responded? */ if (voting_done && !*blocked) { - bit = find_next_bit(mle->maybe_map, O2NM_MAX_NODES, 0); + bit = find_first_bit(mle->maybe_map, O2NM_MAX_NODES); if (dlm->node_num <= bit) { /* my node number is lowest. * now tell other nodes that I am @@ -1234,8 +1234,8 @@ static int dlm_restart_lock_mastery(struct dlm_ctxt *dlm, } else { mlog(ML_ERROR, "node down! %d\n", node); if (blocked) { - int lowest = find_next_bit(mle->maybe_map, - O2NM_MAX_NODES, 0); + int lowest = find_first_bit(mle->maybe_map, + O2NM_MAX_NODES); /* act like it was never there */ clear_bit(node, mle->maybe_map); @@ -1795,7 +1795,7 @@ int dlm_assert_master_handler(struct o2net_msg *msg, u32 len, void *data, "MLE for it! (%.*s)\n", assert->node_idx, namelen, name); } else { - int bit = find_next_bit (mle->maybe_map, O2NM_MAX_NODES, 0); + int bit = find_first_bit(mle->maybe_map, O2NM_MAX_NODES); if (bit >= O2NM_MAX_NODES) { /* not necessarily an error, though less likely. * could be master just re-asserting. */ @@ -2521,7 +2521,7 @@ static int dlm_is_lockres_migratable(struct dlm_ctxt *dlm, } if (!nonlocal) { - node_ref = find_next_bit(res->refmap, O2NM_MAX_NODES, 0); + node_ref = find_first_bit(res->refmap, O2NM_MAX_NODES); if (node_ref >= O2NM_MAX_NODES) return 0; } @@ -3303,7 +3303,7 @@ static void dlm_clean_block_mle(struct dlm_ctxt *dlm, BUG_ON(mle->type != DLM_MLE_BLOCK); spin_lock(&mle->spinlock); - bit = find_next_bit(mle->maybe_map, O2NM_MAX_NODES, 0); + bit = find_first_bit(mle->maybe_map, O2NM_MAX_NODES); if (bit != dead_node) { mlog(0, "mle found, but dead node %u would not have been " "master\n", dead_node); @@ -3542,7 +3542,7 @@ void dlm_force_free_mles(struct dlm_ctxt *dlm) spin_lock(&dlm->master_lock); BUG_ON(dlm->dlm_state != DLM_CTXT_LEAVING); - BUG_ON((find_next_bit(dlm->domain_map, O2NM_MAX_NODES, 0) < O2NM_MAX_NODES)); + BUG_ON((find_first_bit(dlm->domain_map, O2NM_MAX_NODES) < O2NM_MAX_NODES)); for (i = 0; i < DLM_HASH_BUCKETS; i++) { bucket = dlm_master_hash(dlm, i); diff --git a/fs/ocfs2/dlm/dlmrecovery.c b/fs/ocfs2/dlm/dlmrecovery.c index 5cd5f7511dac..52ad342fec3e 100644 --- a/fs/ocfs2/dlm/dlmrecovery.c +++ b/fs/ocfs2/dlm/dlmrecovery.c @@ -451,7 +451,7 @@ static int dlm_do_recovery(struct dlm_ctxt *dlm) if (dlm->reco.dead_node == O2NM_INVALID_NODE_NUM) { int bit; - bit = find_next_bit (dlm->recovery_map, O2NM_MAX_NODES, 0); + bit = find_first_bit(dlm->recovery_map, O2NM_MAX_NODES); if (bit >= O2NM_MAX_NODES || bit < 0) dlm_set_reco_dead_node(dlm, O2NM_INVALID_NODE_NUM); else diff --git a/fs/ocfs2/dlm/dlmthread.c b/fs/ocfs2/dlm/dlmthread.c index c350bd4df770..eedf07ca23ca 100644 --- a/fs/ocfs2/dlm/dlmthread.c +++ b/fs/ocfs2/dlm/dlmthread.c @@ -92,7 +92,7 @@ int __dlm_lockres_unused(struct dlm_lock_resource *res) return 0; /* Another node has this resource with this node as the master */ - bit = find_next_bit(res->refmap, O2NM_MAX_NODES, 0); + bit = find_first_bit(res->refmap, O2NM_MAX_NODES); if (bit < O2NM_MAX_NODES) return 0; diff --git a/fs/ocfs2/filecheck.c b/fs/ocfs2/filecheck.c index de56e6231af8..1ad7106741f8 100644 --- a/fs/ocfs2/filecheck.c +++ b/fs/ocfs2/filecheck.c @@ -94,6 +94,7 @@ static struct attribute *ocfs2_filecheck_attrs[] = { &ocfs2_filecheck_attr_set.attr, NULL }; +ATTRIBUTE_GROUPS(ocfs2_filecheck); static void ocfs2_filecheck_release(struct kobject *kobj) { @@ -138,7 +139,7 @@ static const struct sysfs_ops ocfs2_filecheck_ops = { }; static struct kobj_type ocfs2_ktype_filecheck = { - .default_attrs = ocfs2_filecheck_attrs, + .default_groups = ocfs2_filecheck_groups, .sysfs_ops = &ocfs2_filecheck_ops, .release = ocfs2_filecheck_release, }; diff --git a/fs/ocfs2/journal.c b/fs/ocfs2/journal.c index dbf9b9e97d74..1887a2708709 100644 --- a/fs/ocfs2/journal.c +++ b/fs/ocfs2/journal.c @@ -1669,8 +1669,7 @@ static int ocfs2_replay_journal(struct ocfs2_super *osb, status = jbd2_journal_load(journal); if (status < 0) { mlog_errno(status); - if (!igrab(inode)) - BUG(); + BUG_ON(!igrab(inode)); jbd2_journal_destroy(journal); goto done; } @@ -1699,8 +1698,7 @@ static int ocfs2_replay_journal(struct ocfs2_super *osb, if (status < 0) mlog_errno(status); - if (!igrab(inode)) - BUG(); + BUG_ON(!igrab(inode)); jbd2_journal_destroy(journal); diff --git a/fs/ocfs2/stackglue.c b/fs/ocfs2/stackglue.c index 16f1bfc407f2..731558a6f27d 100644 --- a/fs/ocfs2/stackglue.c +++ b/fs/ocfs2/stackglue.c @@ -672,31 +672,8 @@ static struct ctl_table ocfs2_mod_table[] = { { } }; -static struct ctl_table ocfs2_kern_table[] = { - { - .procname = "ocfs2", - .data = NULL, - .maxlen = 0, - .mode = 0555, - .child = ocfs2_mod_table - }, - { } -}; - -static struct ctl_table ocfs2_root_table[] = { - { - .procname = "fs", - .data = NULL, - .maxlen = 0, - .mode = 0555, - .child = ocfs2_kern_table - }, - { } -}; - static struct ctl_table_header *ocfs2_table_header; - /* * Initialization */ @@ -705,7 +682,7 @@ static int __init ocfs2_stack_glue_init(void) { strcpy(cluster_stack_name, OCFS2_STACK_PLUGIN_O2CB); - ocfs2_table_header = register_sysctl_table(ocfs2_root_table); + ocfs2_table_header = register_sysctl("fs/ocfs2", ocfs2_mod_table); if (!ocfs2_table_header) { printk(KERN_ERR "ocfs2 stack glue: unable to register sysctl\n"); diff --git a/fs/ocfs2/super.c b/fs/ocfs2/super.c index 1286b88b6fa1..2772dec9dcea 100644 --- a/fs/ocfs2/super.c +++ b/fs/ocfs2/super.c @@ -25,7 +25,6 @@ #include <linux/mount.h> #include <linux/seq_file.h> #include <linux/quotaops.h> -#include <linux/cleancache.h> #include <linux/signal.h> #define CREATE_TRACE_POINTS @@ -2283,7 +2282,6 @@ static int ocfs2_initialize_super(struct super_block *sb, mlog_errno(status); goto bail; } - cleancache_init_shared_fs(sb); osb->ocfs2_wq = alloc_ordered_workqueue("ocfs2_wq", WQ_MEM_RECLAIM); if (!osb->ocfs2_wq) { diff --git a/fs/orangefs/orangefs-bufmap.c b/fs/orangefs/orangefs-bufmap.c index 538e839590ef..b501dc07f922 100644 --- a/fs/orangefs/orangefs-bufmap.c +++ b/fs/orangefs/orangefs-bufmap.c @@ -176,7 +176,7 @@ orangefs_bufmap_free(struct orangefs_bufmap *bufmap) { kfree(bufmap->page_array); kfree(bufmap->desc_array); - kfree(bufmap->buffer_index_array); + bitmap_free(bufmap->buffer_index_array); kfree(bufmap); } @@ -226,8 +226,7 @@ orangefs_bufmap_alloc(struct ORANGEFS_dev_map_desc *user_desc) bufmap->desc_size = user_desc->size; bufmap->desc_shift = ilog2(bufmap->desc_size); - bufmap->buffer_index_array = - kzalloc(DIV_ROUND_UP(bufmap->desc_count, BITS_PER_LONG), GFP_KERNEL); + bufmap->buffer_index_array = bitmap_zalloc(bufmap->desc_count, GFP_KERNEL); if (!bufmap->buffer_index_array) goto out_free_bufmap; @@ -250,7 +249,7 @@ orangefs_bufmap_alloc(struct ORANGEFS_dev_map_desc *user_desc) out_free_desc_array: kfree(bufmap->desc_array); out_free_index_array: - kfree(bufmap->buffer_index_array); + bitmap_free(bufmap->buffer_index_array); out_free_bufmap: kfree(bufmap); out: diff --git a/fs/orangefs/orangefs-sysfs.c b/fs/orangefs/orangefs-sysfs.c index 3627ea946402..de80b62553bb 100644 --- a/fs/orangefs/orangefs-sysfs.c +++ b/fs/orangefs/orangefs-sysfs.c @@ -894,10 +894,11 @@ static struct attribute *orangefs_default_attrs[] = { &perf_time_interval_secs_attribute.attr, NULL, }; +ATTRIBUTE_GROUPS(orangefs_default); static struct kobj_type orangefs_ktype = { .sysfs_ops = &orangefs_sysfs_ops, - .default_attrs = orangefs_default_attrs, + .default_groups = orangefs_default_groups, }; static struct orangefs_attribute acache_hard_limit_attribute = @@ -931,10 +932,11 @@ static struct attribute *acache_orangefs_default_attrs[] = { &acache_timeout_msecs_attribute.attr, NULL, }; +ATTRIBUTE_GROUPS(acache_orangefs_default); static struct kobj_type acache_orangefs_ktype = { .sysfs_ops = &orangefs_sysfs_ops, - .default_attrs = acache_orangefs_default_attrs, + .default_groups = acache_orangefs_default_groups, }; static struct orangefs_attribute capcache_hard_limit_attribute = @@ -968,10 +970,11 @@ static struct attribute *capcache_orangefs_default_attrs[] = { &capcache_timeout_secs_attribute.attr, NULL, }; +ATTRIBUTE_GROUPS(capcache_orangefs_default); static struct kobj_type capcache_orangefs_ktype = { .sysfs_ops = &orangefs_sysfs_ops, - .default_attrs = capcache_orangefs_default_attrs, + .default_groups = capcache_orangefs_default_groups, }; static struct orangefs_attribute ccache_hard_limit_attribute = @@ -1005,10 +1008,11 @@ static struct attribute *ccache_orangefs_default_attrs[] = { &ccache_timeout_secs_attribute.attr, NULL, }; +ATTRIBUTE_GROUPS(ccache_orangefs_default); static struct kobj_type ccache_orangefs_ktype = { .sysfs_ops = &orangefs_sysfs_ops, - .default_attrs = ccache_orangefs_default_attrs, + .default_groups = ccache_orangefs_default_groups, }; static struct orangefs_attribute ncache_hard_limit_attribute = @@ -1042,10 +1046,11 @@ static struct attribute *ncache_orangefs_default_attrs[] = { &ncache_timeout_msecs_attribute.attr, NULL, }; +ATTRIBUTE_GROUPS(ncache_orangefs_default); static struct kobj_type ncache_orangefs_ktype = { .sysfs_ops = &orangefs_sysfs_ops, - .default_attrs = ncache_orangefs_default_attrs, + .default_groups = ncache_orangefs_default_groups, }; static struct orangefs_attribute pc_acache_attribute = @@ -1072,10 +1077,11 @@ static struct attribute *pc_orangefs_default_attrs[] = { &pc_ncache_attribute.attr, NULL, }; +ATTRIBUTE_GROUPS(pc_orangefs_default); static struct kobj_type pc_orangefs_ktype = { .sysfs_ops = &orangefs_sysfs_ops, - .default_attrs = pc_orangefs_default_attrs, + .default_groups = pc_orangefs_default_groups, }; static struct orangefs_attribute stats_reads_attribute = @@ -1095,10 +1101,11 @@ static struct attribute *stats_orangefs_default_attrs[] = { &stats_writes_attribute.attr, NULL, }; +ATTRIBUTE_GROUPS(stats_orangefs_default); static struct kobj_type stats_orangefs_ktype = { .sysfs_ops = &orangefs_sysfs_ops, - .default_attrs = stats_orangefs_default_attrs, + .default_groups = stats_orangefs_default_groups, }; static struct kobject *orangefs_obj; diff --git a/fs/pipe.c b/fs/pipe.c index 6d4342bad9f1..cc28623a67b6 100644 --- a/fs/pipe.c +++ b/fs/pipe.c @@ -25,6 +25,7 @@ #include <linux/fcntl.h> #include <linux/memcontrol.h> #include <linux/watch_queue.h> +#include <linux/sysctl.h> #include <linux/uaccess.h> #include <asm/ioctls.h> @@ -50,13 +51,13 @@ * The max size that a non-root user is allowed to grow the pipe. Can * be set by root in /proc/sys/fs/pipe-max-size */ -unsigned int pipe_max_size = 1048576; +static unsigned int pipe_max_size = 1048576; /* Maximum allocatable pages per user. Hard limit is unset by default, soft * matches default values. */ -unsigned long pipe_user_pages_hard; -unsigned long pipe_user_pages_soft = PIPE_DEF_BUFFERS * INR_OPEN_CUR; +static unsigned long pipe_user_pages_hard; +static unsigned long pipe_user_pages_soft = PIPE_DEF_BUFFERS * INR_OPEN_CUR; /* * We use head and tail indices that aren't masked off, except at the point of @@ -1428,6 +1429,60 @@ static struct file_system_type pipe_fs_type = { .kill_sb = kill_anon_super, }; +#ifdef CONFIG_SYSCTL +static int do_proc_dopipe_max_size_conv(unsigned long *lvalp, + unsigned int *valp, + int write, void *data) +{ + if (write) { + unsigned int val; + + val = round_pipe_size(*lvalp); + if (val == 0) + return -EINVAL; + + *valp = val; + } else { + unsigned int val = *valp; + *lvalp = (unsigned long) val; + } + + return 0; +} + +static int proc_dopipe_max_size(struct ctl_table *table, int write, + void *buffer, size_t *lenp, loff_t *ppos) +{ + return do_proc_douintvec(table, write, buffer, lenp, ppos, + do_proc_dopipe_max_size_conv, NULL); +} + +static struct ctl_table fs_pipe_sysctls[] = { + { + .procname = "pipe-max-size", + .data = &pipe_max_size, + .maxlen = sizeof(pipe_max_size), + .mode = 0644, + .proc_handler = proc_dopipe_max_size, + }, + { + .procname = "pipe-user-pages-hard", + .data = &pipe_user_pages_hard, + .maxlen = sizeof(pipe_user_pages_hard), + .mode = 0644, + .proc_handler = proc_doulongvec_minmax, + }, + { + .procname = "pipe-user-pages-soft", + .data = &pipe_user_pages_soft, + .maxlen = sizeof(pipe_user_pages_soft), + .mode = 0644, + .proc_handler = proc_doulongvec_minmax, + }, + { } +}; +#endif + static int __init init_pipe_fs(void) { int err = register_filesystem(&pipe_fs_type); @@ -1439,6 +1494,9 @@ static int __init init_pipe_fs(void) unregister_filesystem(&pipe_fs_type); } } +#ifdef CONFIG_SYSCTL + register_sysctl_init("fs", fs_pipe_sysctls); +#endif return err; } diff --git a/fs/proc/array.c b/fs/proc/array.c index ff869a66b34e..fd8b0c12b2cb 100644 --- a/fs/proc/array.c +++ b/fs/proc/array.c @@ -92,6 +92,7 @@ #include <linux/string_helpers.h> #include <linux/user_namespace.h> #include <linux/fs_struct.h> +#include <linux/kthread.h> #include <asm/processor.h> #include "internal.h" @@ -102,6 +103,8 @@ void proc_task_name(struct seq_file *m, struct task_struct *p, bool escape) if (p->flags & PF_WQ_WORKER) wq_worker_comm(tcomm, sizeof(tcomm), p); + else if (p->flags & PF_KTHREAD) + get_kthread_comm(tcomm, sizeof(tcomm), p); else __get_task_comm(tcomm, sizeof(tcomm), p); @@ -468,6 +471,7 @@ static int do_task_stat(struct seq_file *m, struct pid_namespace *ns, u64 cgtime, gtime; unsigned long rsslim = 0; unsigned long flags; + int exit_code = task->exit_code; state = *get_task_state(task); vsize = eip = esp = 0; @@ -531,6 +535,9 @@ static int do_task_stat(struct seq_file *m, struct pid_namespace *ns, maj_flt += sig->maj_flt; thread_group_cputime_adjusted(task, &utime, &stime); gtime += sig->gtime; + + if (sig->flags & (SIGNAL_GROUP_EXIT | SIGNAL_STOP_STOPPED)) + exit_code = sig->group_exit_code; } sid = task_session_nr_ns(task, ns); @@ -630,7 +637,7 @@ static int do_task_stat(struct seq_file *m, struct pid_namespace *ns, seq_puts(m, " 0 0 0 0 0 0 0"); if (permitted) - seq_put_decimal_ll(m, " ", task->exit_code); + seq_put_decimal_ll(m, " ", exit_code); else seq_puts(m, " 0"); diff --git a/fs/proc/base.c b/fs/proc/base.c index 13eda8de2998..d654ce7150fd 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -670,10 +670,10 @@ static int proc_pid_syscall(struct seq_file *m, struct pid_namespace *ns, /************************************************************************/ /* permission checks */ -static int proc_fd_access_allowed(struct inode *inode) +static bool proc_fd_access_allowed(struct inode *inode) { struct task_struct *task; - int allowed = 0; + bool allowed = false; /* Allow access to a task's file descriptors if it is us or we * may use ptrace attach to the process and find out that * information. diff --git a/fs/proc/generic.c b/fs/proc/generic.c index 5b78739e60e4..f2132407e133 100644 --- a/fs/proc/generic.c +++ b/fs/proc/generic.c @@ -791,12 +791,6 @@ void proc_remove(struct proc_dir_entry *de) } EXPORT_SYMBOL(proc_remove); -void *PDE_DATA(const struct inode *inode) -{ - return __PDE_DATA(inode); -} -EXPORT_SYMBOL(PDE_DATA); - /* * Pull a user buffer into memory and pass it to the file's write handler if * one is supplied. The ->write() method is permitted to modify the diff --git a/fs/proc/inode.c b/fs/proc/inode.c index 599eb724ff2d..f84355c5a36d 100644 --- a/fs/proc/inode.c +++ b/fs/proc/inode.c @@ -650,6 +650,7 @@ struct inode *proc_get_inode(struct super_block *sb, struct proc_dir_entry *de) return NULL; } + inode->i_private = de->data; inode->i_ino = de->low_ino; inode->i_mtime = inode->i_atime = inode->i_ctime = current_time(inode); PROC_I(inode)->pde = de; diff --git a/fs/proc/internal.h b/fs/proc/internal.h index 03415f3fb3a8..06a80f78433d 100644 --- a/fs/proc/internal.h +++ b/fs/proc/internal.h @@ -115,11 +115,6 @@ static inline struct proc_dir_entry *PDE(const struct inode *inode) return PROC_I(inode)->pde; } -static inline void *__PDE_DATA(const struct inode *inode) -{ - return PDE(inode)->data; -} - static inline struct pid *proc_pid(const struct inode *inode) { return PROC_I(inode)->pid; diff --git a/fs/proc/proc_net.c b/fs/proc/proc_net.c index 39b823ab2564..e1cfeda397f3 100644 --- a/fs/proc/proc_net.c +++ b/fs/proc/proc_net.c @@ -138,7 +138,7 @@ EXPORT_SYMBOL_GPL(proc_create_net_data); * @parent: The parent directory in which to create. * @ops: The seq_file ops with which to read the file. * @write: The write method with which to 'modify' the file. - * @data: Data for retrieval by PDE_DATA(). + * @data: Data for retrieval by pde_data(). * * Create a network namespaced proc file in the @parent directory with the * specified @name and @mode that allows reading of a file that displays a @@ -153,7 +153,7 @@ EXPORT_SYMBOL_GPL(proc_create_net_data); * modified by the @write function. @write should return 0 on success. * * The @data value is accessible from the @show and @write functions by calling - * PDE_DATA() on the file inode. The network namespace must be accessed by + * pde_data() on the file inode. The network namespace must be accessed by * calling seq_file_net() on the seq_file struct. */ struct proc_dir_entry *proc_create_net_data_write(const char *name, umode_t mode, @@ -230,7 +230,7 @@ EXPORT_SYMBOL_GPL(proc_create_net_single); * @parent: The parent directory in which to create. * @show: The seqfile show method with which to read the file. * @write: The write method with which to 'modify' the file. - * @data: Data for retrieval by PDE_DATA(). + * @data: Data for retrieval by pde_data(). * * Create a network-namespaced proc file in the @parent directory with the * specified @name and @mode that allows reading of a file that displays a @@ -245,7 +245,7 @@ EXPORT_SYMBOL_GPL(proc_create_net_single); * modified by the @write function. @write should return 0 on success. * * The @data value is accessible from the @show and @write functions by calling - * PDE_DATA() on the file inode. The network namespace must be accessed by + * pde_data() on the file inode. The network namespace must be accessed by * calling seq_file_single_net() on the seq_file struct. */ struct proc_dir_entry *proc_create_net_single_write(const char *name, umode_t mode, diff --git a/fs/proc/proc_sysctl.c b/fs/proc/proc_sysctl.c index 5d66faecd4ef..7d9cfc730bd4 100644 --- a/fs/proc/proc_sysctl.c +++ b/fs/proc/proc_sysctl.c @@ -16,6 +16,7 @@ #include <linux/module.h> #include <linux/bpf-cgroup.h> #include <linux/mount.h> +#include <linux/kmemleak.h> #include "internal.h" static const struct dentry_operations proc_sys_dentry_operations; @@ -25,15 +26,32 @@ static const struct file_operations proc_sys_dir_file_operations; static const struct inode_operations proc_sys_dir_operations; /* shared constants to be used in various sysctls */ -const int sysctl_vals[] = { 0, 1, INT_MAX }; +const int sysctl_vals[] = { -1, 0, 1, 2, 4, 100, 200, 1000, 3000, INT_MAX, 65535 }; EXPORT_SYMBOL(sysctl_vals); +const unsigned long sysctl_long_vals[] = { 0, 1, LONG_MAX }; +EXPORT_SYMBOL_GPL(sysctl_long_vals); + /* Support for permanently empty directories */ struct ctl_table sysctl_mount_point[] = { { } }; +/** + * register_sysctl_mount_point() - registers a sysctl mount point + * @path: path for the mount point + * + * Used to create a permanently empty directory to serve as mount point. + * There are some subtle but important permission checks this allows in the + * case of unprivileged mounts. + */ +struct ctl_table_header *register_sysctl_mount_point(const char *path) +{ + return register_sysctl(path, sysctl_mount_point); +} +EXPORT_SYMBOL(register_sysctl_mount_point); + static bool is_empty_dir(struct ctl_table_header *head) { return head->ctl_table[0].child == sysctl_mount_point; @@ -163,7 +181,7 @@ static int insert_entry(struct ctl_table_header *head, struct ctl_table *entry) else { pr_err("sysctl duplicate entry: "); sysctl_print_dir(head->parent); - pr_cont("/%s\n", entry->procname); + pr_cont("%s\n", entry->procname); return -EEXIST; } } @@ -1020,8 +1038,8 @@ failed: if (IS_ERR(subdir)) { pr_err("sysctl could not get directory: "); sysctl_print_dir(dir); - pr_cont("/%*.*s %ld\n", - namelen, namelen, name, PTR_ERR(subdir)); + pr_cont("%*.*s %ld\n", namelen, namelen, name, + PTR_ERR(subdir)); } drop_sysctl_table(&dir->header); if (new) @@ -1053,7 +1071,6 @@ static int sysctl_follow_link(struct ctl_table_header **phead, struct ctl_dir *dir; int ret; - ret = 0; spin_lock(&sysctl_lock); root = (*pentry)->data; set = lookup_header_set(root); @@ -1384,6 +1401,38 @@ struct ctl_table_header *register_sysctl(const char *path, struct ctl_table *tab } EXPORT_SYMBOL(register_sysctl); +/** + * __register_sysctl_init() - register sysctl table to path + * @path: path name for sysctl base + * @table: This is the sysctl table that needs to be registered to the path + * @table_name: The name of sysctl table, only used for log printing when + * registration fails + * + * The sysctl interface is used by userspace to query or modify at runtime + * a predefined value set on a variable. These variables however have default + * values pre-set. Code which depends on these variables will always work even + * if register_sysctl() fails. If register_sysctl() fails you'd just loose the + * ability to query or modify the sysctls dynamically at run time. Chances of + * register_sysctl() failing on init are extremely low, and so for both reasons + * this function does not return any error as it is used by initialization code. + * + * Context: Can only be called after your respective sysctl base path has been + * registered. So for instance, most base directories are registered early on + * init before init levels are processed through proc_sys_init() and + * sysctl_init_bases(). + */ +void __init __register_sysctl_init(const char *path, struct ctl_table *table, + const char *table_name) +{ + struct ctl_table_header *hdr = register_sysctl(path, table); + + if (unlikely(!hdr)) { + pr_err("failed when register_sysctl %s to %s\n", table_name, path); + return; + } + kmemleak_not_leak(hdr); +} + static char *append_path(const char *path, char *pos, const char *name) { int namelen; @@ -1597,6 +1646,15 @@ struct ctl_table_header *register_sysctl_table(struct ctl_table *table) } EXPORT_SYMBOL(register_sysctl_table); +int __register_sysctl_base(struct ctl_table *base_table) +{ + struct ctl_table_header *hdr; + + hdr = register_sysctl_table(base_table); + kmemleak_not_leak(hdr); + return 0; +} + static void put_links(struct ctl_table_header *header) { struct ctl_table_set *root_set = &sysctl_table_root.default_set; @@ -1626,7 +1684,7 @@ static void put_links(struct ctl_table_header *header) else { pr_err("sysctl link missing during unregister: "); sysctl_print_dir(parent); - pr_cont("/%s\n", name); + pr_cont("%s\n", name); } } } @@ -1710,7 +1768,7 @@ int __init proc_sys_init(void) proc_sys_root->proc_dir_ops = &proc_sys_dir_file_operations; proc_sys_root->nlink = 0; - return sysctl_init(); + return sysctl_init_bases(); } struct sysctl_alias { diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c index ad667dbc96f5..18f8c3acbb85 100644 --- a/fs/proc/task_mmu.c +++ b/fs/proc/task_mmu.c @@ -1,6 +1,7 @@ // SPDX-License-Identifier: GPL-2.0 #include <linux/pagewalk.h> #include <linux/vmacache.h> +#include <linux/mm_inline.h> #include <linux/hugetlb.h> #include <linux/huge_mm.h> #include <linux/mount.h> @@ -308,6 +309,8 @@ show_map_vma(struct seq_file *m, struct vm_area_struct *vma) name = arch_vma_name(vma); if (!name) { + const char *anon_name; + if (!mm) { name = "[vdso]"; goto done; @@ -319,8 +322,16 @@ show_map_vma(struct seq_file *m, struct vm_area_struct *vma) goto done; } - if (is_stack(vma)) + if (is_stack(vma)) { name = "[stack]"; + goto done; + } + + anon_name = vma_anon_name(vma); + if (anon_name) { + seq_pad(m, ' '); + seq_printf(m, "[anon:%s]", anon_name); + } } done: diff --git a/fs/proc/vmcore.c b/fs/proc/vmcore.c index 509f85148fee..702754dd1daf 100644 --- a/fs/proc/vmcore.c +++ b/fs/proc/vmcore.c @@ -65,8 +65,6 @@ static size_t vmcoredd_orig_sz; static DECLARE_RWSEM(vmcore_cb_rwsem); /* List of registered vmcore callbacks. */ static LIST_HEAD(vmcore_cb_list); -/* Whether we had a surprise unregistration of a callback. */ -static bool vmcore_cb_unstable; /* Whether the vmcore has been opened once. */ static bool vmcore_opened; @@ -94,10 +92,8 @@ void unregister_vmcore_cb(struct vmcore_cb *cb) * very unusual (e.g., forced driver removal), but we cannot stop * unregistering. */ - if (vmcore_opened) { + if (vmcore_opened) pr_warn_once("Unexpected vmcore callback unregistration\n"); - vmcore_cb_unstable = true; - } up_write(&vmcore_cb_rwsem); } EXPORT_SYMBOL_GPL(unregister_vmcore_cb); @@ -108,8 +104,6 @@ static bool pfn_is_ram(unsigned long pfn) bool ret = true; lockdep_assert_held_read(&vmcore_cb_rwsem); - if (unlikely(vmcore_cb_unstable)) - return false; list_for_each_entry(cb, &vmcore_cb_list, next) { if (unlikely(!cb->pfn_is_ram)) @@ -581,7 +575,7 @@ static int vmcore_remap_oldmem_pfn(struct vm_area_struct *vma, * looping over all pages without a reason. */ down_read(&vmcore_cb_rwsem); - if (!list_empty(&vmcore_cb_list) || vmcore_cb_unstable) + if (!list_empty(&vmcore_cb_list)) ret = remap_oldmem_pfn_checked(vma, from, pfn, size, prot); else ret = remap_oldmem_pfn_range(vma, from, pfn, size, prot); diff --git a/fs/signalfd.c b/fs/signalfd.c index 65ce0e72e7b9..e20d1484c663 100644 --- a/fs/signalfd.c +++ b/fs/signalfd.c @@ -155,11 +155,12 @@ static int signalfd_copyinfo(struct signalfd_siginfo __user *uinfo, static ssize_t signalfd_dequeue(struct signalfd_ctx *ctx, kernel_siginfo_t *info, int nonblock) { + enum pid_type type; ssize_t ret; DECLARE_WAITQUEUE(wait, current); spin_lock_irq(¤t->sighand->siglock); - ret = dequeue_signal(current, &ctx->sigmask, info); + ret = dequeue_signal(current, &ctx->sigmask, info, &type); switch (ret) { case 0: if (!nonblock) @@ -174,7 +175,7 @@ static ssize_t signalfd_dequeue(struct signalfd_ctx *ctx, kernel_siginfo_t *info add_wait_queue(¤t->sighand->signalfd_wqh, &wait); for (;;) { set_current_state(TASK_INTERRUPTIBLE); - ret = dequeue_signal(current, &ctx->sigmask, info); + ret = dequeue_signal(current, &ctx->sigmask, info, &type); if (ret != 0) break; if (signal_pending(current)) { diff --git a/fs/smbfs_common/smb2pdu.h b/fs/smbfs_common/smb2pdu.h index 7ccadcbe684b..38b8fc514860 100644 --- a/fs/smbfs_common/smb2pdu.h +++ b/fs/smbfs_common/smb2pdu.h @@ -449,7 +449,7 @@ struct smb2_netname_neg_context { */ /* Flags */ -#define SMB2_ACCEPT_TRANSFORM_LEVEL_SECURITY 0x00000001 +#define SMB2_ACCEPT_TRANSPORT_LEVEL_SECURITY 0x00000001 struct smb2_transport_capabilities_context { __le16 ContextType; /* 6 */ diff --git a/fs/smbfs_common/smbfsctl.h b/fs/smbfs_common/smbfsctl.h index 926f87cd6af0..d51939c43ad7 100644 --- a/fs/smbfs_common/smbfsctl.h +++ b/fs/smbfs_common/smbfsctl.h @@ -95,8 +95,10 @@ #define FSCTL_SET_SHORT_NAME_BEHAVIOR 0x000901B4 /* BB add struct */ #define FSCTL_GET_INTEGRITY_INFORMATION 0x0009027C #define FSCTL_GET_REFS_VOLUME_DATA 0x000902D8 /* See MS-FSCC 2.3.24 */ +#define FSCTL_SET_INTEGRITY_INFORMATION_EXT 0x00090380 #define FSCTL_GET_RETRIEVAL_POINTERS_AND_REFCOUNT 0x000903d3 #define FSCTL_GET_RETRIEVAL_POINTER_COUNT 0x0009042b +#define FSCTL_REFS_STREAM_SNAPSHOT_MANAGEMENT 0x00090440 #define FSCTL_QUERY_ALLOCATED_RANGES 0x000940CF #define FSCTL_SET_DEFECT_MANAGEMENT 0x00098134 /* BB add struct */ #define FSCTL_FILE_LEVEL_TRIM 0x00098208 /* BB add struct */ diff --git a/fs/squashfs/super.c b/fs/squashfs/super.c index bb44ff4c5cc6..b1b556dbce12 100644 --- a/fs/squashfs/super.c +++ b/fs/squashfs/super.c @@ -29,6 +29,7 @@ #include <linux/module.h> #include <linux/magic.h> #include <linux/xattr.h> +#include <linux/backing-dev.h> #include "squashfs_fs.h" #include "squashfs_fs_sb.h" @@ -112,6 +113,24 @@ static const struct squashfs_decompressor *supported_squashfs_filesystem( return decompressor; } +static int squashfs_bdi_init(struct super_block *sb) +{ + int err; + unsigned int major = MAJOR(sb->s_dev); + unsigned int minor = MINOR(sb->s_dev); + + bdi_put(sb->s_bdi); + sb->s_bdi = &noop_backing_dev_info; + + err = super_setup_bdi_name(sb, "squashfs_%u_%u", major, minor); + if (err) + return err; + + sb->s_bdi->ra_pages = 0; + sb->s_bdi->io_pages = 0; + + return 0; +} static int squashfs_fill_super(struct super_block *sb, struct fs_context *fc) { @@ -127,6 +146,20 @@ static int squashfs_fill_super(struct super_block *sb, struct fs_context *fc) TRACE("Entered squashfs_fill_superblock\n"); + /* + * squashfs provides 'backing_dev_info' in order to disable read-ahead. For + * squashfs, I/O is not deferred, it is done immediately in readpage, + * which means the user would always have to wait their own I/O. So the effect + * of readahead is very weak for squashfs. squashfs_bdi_init will set + * sb->s_bdi->ra_pages and sb->s_bdi->io_pages to 0 and close readahead for + * squashfs. + */ + err = squashfs_bdi_init(sb); + if (err) { + errorf(fc, "squashfs init bdi failed"); + return err; + } + sb->s_fs_info = kzalloc(sizeof(*msblk), GFP_KERNEL); if (sb->s_fs_info == NULL) { ERROR("Failed to allocate squashfs_sb_info\n"); diff --git a/fs/super.c b/fs/super.c index 3bfc0f8fbd5b..7af820ba5ad5 100644 --- a/fs/super.c +++ b/fs/super.c @@ -31,7 +31,6 @@ #include <linux/mutex.h> #include <linux/backing-dev.h> #include <linux/rculist_bl.h> -#include <linux/cleancache.h> #include <linux/fscrypt.h> #include <linux/fsnotify.h> #include <linux/lockdep.h> @@ -260,7 +259,6 @@ static struct super_block *alloc_super(struct file_system_type *type, int flags, s->s_time_gran = 1000000000; s->s_time_min = TIME64_MIN; s->s_time_max = TIME64_MAX; - s->cleancache_poolid = CLEANCACHE_NO_POOL; s->s_shrink.seeks = DEFAULT_SEEKS; s->s_shrink.scan_objects = super_cache_scan; @@ -330,7 +328,6 @@ void deactivate_locked_super(struct super_block *s) { struct file_system_type *fs = s->s_type; if (atomic_dec_and_test(&s->s_active)) { - cleancache_invalidate_fs(s); unregister_shrinker(&s->s_shrink); fs->kill_sb(s); @@ -1423,8 +1420,8 @@ struct dentry *mount_nodev(struct file_system_type *fs_type, } EXPORT_SYMBOL(mount_nodev); -static int reconfigure_single(struct super_block *s, - int flags, void *data) +int reconfigure_single(struct super_block *s, + int flags, void *data) { struct fs_context *fc; int ret; diff --git a/fs/sysctls.c b/fs/sysctls.c new file mode 100644 index 000000000000..c701273c9432 --- /dev/null +++ b/fs/sysctls.c @@ -0,0 +1,39 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * /proc/sys/fs shared sysctls + * + * These sysctls are shared between different filesystems. + */ +#include <linux/init.h> +#include <linux/sysctl.h> + +static struct ctl_table fs_shared_sysctls[] = { + { + .procname = "overflowuid", + .data = &fs_overflowuid, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = SYSCTL_ZERO, + .extra2 = SYSCTL_MAXOLDUID, + }, + { + .procname = "overflowgid", + .data = &fs_overflowgid, + .maxlen = sizeof(int), + .mode = 0644, + .proc_handler = proc_dointvec_minmax, + .extra1 = SYSCTL_ZERO, + .extra2 = SYSCTL_MAXOLDUID, + }, + { } +}; + +DECLARE_SYSCTL_BASE(fs, fs_shared_sysctls); + +static int __init init_fs_sysctls(void) +{ + return register_sysctl_base(fs); +} + +early_initcall(init_fs_sysctls); diff --git a/fs/tracefs/inode.c b/fs/tracefs/inode.c index 3616839c5c4b..bafc02bf8220 100644 --- a/fs/tracefs/inode.c +++ b/fs/tracefs/inode.c @@ -109,12 +109,12 @@ static int tracefs_syscall_rmdir(struct inode *inode, struct dentry *dentry) * also the directory that is being deleted. */ inode_unlock(inode); - inode_unlock(dentry->d_inode); + inode_unlock(d_inode(dentry)); ret = tracefs_ops.rmdir(name); inode_lock_nested(inode, I_MUTEX_PARENT); - inode_lock(dentry->d_inode); + inode_lock(d_inode(dentry)); kfree(name); @@ -284,7 +284,7 @@ static int tracefs_parse_options(char *data, struct tracefs_mount_opts *opts) static int tracefs_apply_options(struct super_block *sb) { struct tracefs_fs_info *fsi = sb->s_fs_info; - struct inode *inode = sb->s_root->d_inode; + struct inode *inode = d_inode(sb->s_root); struct tracefs_mount_opts *opts = &fsi->mount_opts; inode->i_mode &= ~S_IALLUGO; @@ -403,18 +403,18 @@ static struct dentry *start_creating(const char *name, struct dentry *parent) if (!parent) parent = tracefs_mount->mnt_root; - inode_lock(parent->d_inode); - if (unlikely(IS_DEADDIR(parent->d_inode))) + inode_lock(d_inode(parent)); + if (unlikely(IS_DEADDIR(d_inode(parent)))) dentry = ERR_PTR(-ENOENT); else dentry = lookup_one_len(name, parent, strlen(name)); - if (!IS_ERR(dentry) && dentry->d_inode) { + if (!IS_ERR(dentry) && d_inode(dentry)) { dput(dentry); dentry = ERR_PTR(-EEXIST); } if (IS_ERR(dentry)) { - inode_unlock(parent->d_inode); + inode_unlock(d_inode(parent)); simple_release_fs(&tracefs_mount, &tracefs_mount_count); } @@ -423,7 +423,7 @@ static struct dentry *start_creating(const char *name, struct dentry *parent) static struct dentry *failed_creating(struct dentry *dentry) { - inode_unlock(dentry->d_parent->d_inode); + inode_unlock(d_inode(dentry->d_parent)); dput(dentry); simple_release_fs(&tracefs_mount, &tracefs_mount_count); return NULL; @@ -431,7 +431,7 @@ static struct dentry *failed_creating(struct dentry *dentry) static struct dentry *end_creating(struct dentry *dentry) { - inode_unlock(dentry->d_parent->d_inode); + inode_unlock(d_inode(dentry->d_parent)); return dentry; } @@ -489,7 +489,7 @@ struct dentry *tracefs_create_file(const char *name, umode_t mode, inode->i_uid = d_inode(dentry->d_parent)->i_uid; inode->i_gid = d_inode(dentry->d_parent)->i_gid; d_instantiate(dentry, inode); - fsnotify_create(dentry->d_parent->d_inode, dentry); + fsnotify_create(d_inode(dentry->d_parent), dentry); return end_creating(dentry); } @@ -516,8 +516,8 @@ static struct dentry *__create_dir(const char *name, struct dentry *parent, /* directory inodes start off with i_nlink == 2 (for "." entry) */ inc_nlink(inode); d_instantiate(dentry, inode); - inc_nlink(dentry->d_parent->d_inode); - fsnotify_mkdir(dentry->d_parent->d_inode, dentry); + inc_nlink(d_inode(dentry->d_parent)); + fsnotify_mkdir(d_inode(dentry->d_parent), dentry); return end_creating(dentry); } diff --git a/fs/unicode/.gitignore b/fs/unicode/.gitignore index 361294571ab0..51cdf3fb4dd4 100644 --- a/fs/unicode/.gitignore +++ b/fs/unicode/.gitignore @@ -1,3 +1,3 @@ # SPDX-License-Identifier: GPL-2.0-only /mkutf8data -/utf8data.h +/utf8data.c diff --git a/fs/unicode/Kconfig b/fs/unicode/Kconfig index 2c27b9a5cd6c..610d7bc05d6e 100644 --- a/fs/unicode/Kconfig +++ b/fs/unicode/Kconfig @@ -8,7 +8,16 @@ config UNICODE Say Y here to enable UTF-8 NFD normalization and NFD+CF casefolding support. +config UNICODE_UTF8_DATA + tristate "UTF-8 normalization and casefolding tables" + depends on UNICODE + default UNICODE + help + This contains a large table of case foldings, which can be loaded as + a separate module if you say M here. To be on the safe side stick + to the default of Y. Saying N here makes no sense, if you do not want + utf8 casefolding support, disable CONFIG_UNICODE instead. + config UNICODE_NORMALIZATION_SELFTEST tristate "Test UTF-8 normalization support" - depends on UNICODE - default n + depends on UNICODE_UTF8_DATA diff --git a/fs/unicode/Makefile b/fs/unicode/Makefile index b88aecc86550..2f9d9188852b 100644 --- a/fs/unicode/Makefile +++ b/fs/unicode/Makefile @@ -2,14 +2,15 @@ obj-$(CONFIG_UNICODE) += unicode.o obj-$(CONFIG_UNICODE_NORMALIZATION_SELFTEST) += utf8-selftest.o +obj-$(CONFIG_UNICODE_UTF8_DATA) += utf8data.o unicode-y := utf8-norm.o utf8-core.o -$(obj)/utf8-norm.o: $(obj)/utf8data.h +$(obj)/utf8-data.o: $(obj)/utf8data.c -# In the normal build, the checked-in utf8data.h is just shipped. +# In the normal build, the checked-in utf8data.c is just shipped. # -# To generate utf8data.h from UCD, put *.txt files in this directory +# To generate utf8data.c from UCD, put *.txt files in this directory # and pass REGENERATE_UTF8DATA=1 from the command line. ifdef REGENERATE_UTF8DATA @@ -24,15 +25,15 @@ quiet_cmd_utf8data = GEN $@ -t $(srctree)/$(src)/NormalizationTest.txt \ -o $@ -$(obj)/utf8data.h: $(obj)/mkutf8data $(filter %.txt, $(cmd_utf8data)) FORCE +$(obj)/utf8data.c: $(obj)/mkutf8data $(filter %.txt, $(cmd_utf8data)) FORCE $(call if_changed,utf8data) else -$(obj)/utf8data.h: $(src)/utf8data.h_shipped FORCE +$(obj)/utf8data.c: $(src)/utf8data.c_shipped FORCE $(call if_changed,shipped) endif -targets += utf8data.h +targets += utf8data.c hostprogs += mkutf8data diff --git a/fs/unicode/mkutf8data.c b/fs/unicode/mkutf8data.c index ff2025ac5a32..bc1a7c8b5c8d 100644 --- a/fs/unicode/mkutf8data.c +++ b/fs/unicode/mkutf8data.c @@ -3287,12 +3287,10 @@ static void write_file(void) open_fail(utf8_name, errno); fprintf(file, "/* This file is generated code, do not edit. */\n"); - fprintf(file, "#ifndef __INCLUDED_FROM_UTF8NORM_C__\n"); - fprintf(file, "#error Only nls_utf8-norm.c should include this file.\n"); - fprintf(file, "#endif\n"); fprintf(file, "\n"); - fprintf(file, "static const unsigned int utf8vers = %#x;\n", - unicode_maxage); + fprintf(file, "#include <linux/module.h>\n"); + fprintf(file, "#include <linux/kernel.h>\n"); + fprintf(file, "#include \"utf8n.h\"\n"); fprintf(file, "\n"); fprintf(file, "static const unsigned int utf8agetab[] = {\n"); for (i = 0; i != ages_count; i++) @@ -3339,6 +3337,22 @@ static void write_file(void) fprintf(file, "\n"); } fprintf(file, "};\n"); + fprintf(file, "\n"); + fprintf(file, "struct utf8data_table utf8_data_table = {\n"); + fprintf(file, "\t.utf8agetab = utf8agetab,\n"); + fprintf(file, "\t.utf8agetab_size = ARRAY_SIZE(utf8agetab),\n"); + fprintf(file, "\n"); + fprintf(file, "\t.utf8nfdicfdata = utf8nfdicfdata,\n"); + fprintf(file, "\t.utf8nfdicfdata_size = ARRAY_SIZE(utf8nfdicfdata),\n"); + fprintf(file, "\n"); + fprintf(file, "\t.utf8nfdidata = utf8nfdidata,\n"); + fprintf(file, "\t.utf8nfdidata_size = ARRAY_SIZE(utf8nfdidata),\n"); + fprintf(file, "\n"); + fprintf(file, "\t.utf8data = utf8data,\n"); + fprintf(file, "};\n"); + fprintf(file, "EXPORT_SYMBOL_GPL(utf8_data_table);"); + fprintf(file, "\n"); + fprintf(file, "MODULE_LICENSE(\"GPL v2\");\n"); fclose(file); } diff --git a/fs/unicode/utf8-core.c b/fs/unicode/utf8-core.c index dc25823bfed9..67aaadc3ab07 100644 --- a/fs/unicode/utf8-core.c +++ b/fs/unicode/utf8-core.c @@ -5,16 +5,13 @@ #include <linux/slab.h> #include <linux/parser.h> #include <linux/errno.h> -#include <linux/unicode.h> #include <linux/stringhash.h> #include "utf8n.h" int utf8_validate(const struct unicode_map *um, const struct qstr *str) { - const struct utf8data *data = utf8nfdi(um->version); - - if (utf8nlen(data, str->name, str->len) < 0) + if (utf8nlen(um, UTF8_NFDI, str->name, str->len) < 0) return -1; return 0; } @@ -23,14 +20,13 @@ EXPORT_SYMBOL(utf8_validate); int utf8_strncmp(const struct unicode_map *um, const struct qstr *s1, const struct qstr *s2) { - const struct utf8data *data = utf8nfdi(um->version); struct utf8cursor cur1, cur2; int c1, c2; - if (utf8ncursor(&cur1, data, s1->name, s1->len) < 0) + if (utf8ncursor(&cur1, um, UTF8_NFDI, s1->name, s1->len) < 0) return -EINVAL; - if (utf8ncursor(&cur2, data, s2->name, s2->len) < 0) + if (utf8ncursor(&cur2, um, UTF8_NFDI, s2->name, s2->len) < 0) return -EINVAL; do { @@ -50,14 +46,13 @@ EXPORT_SYMBOL(utf8_strncmp); int utf8_strncasecmp(const struct unicode_map *um, const struct qstr *s1, const struct qstr *s2) { - const struct utf8data *data = utf8nfdicf(um->version); struct utf8cursor cur1, cur2; int c1, c2; - if (utf8ncursor(&cur1, data, s1->name, s1->len) < 0) + if (utf8ncursor(&cur1, um, UTF8_NFDICF, s1->name, s1->len) < 0) return -EINVAL; - if (utf8ncursor(&cur2, data, s2->name, s2->len) < 0) + if (utf8ncursor(&cur2, um, UTF8_NFDICF, s2->name, s2->len) < 0) return -EINVAL; do { @@ -81,12 +76,11 @@ int utf8_strncasecmp_folded(const struct unicode_map *um, const struct qstr *cf, const struct qstr *s1) { - const struct utf8data *data = utf8nfdicf(um->version); struct utf8cursor cur1; int c1, c2; int i = 0; - if (utf8ncursor(&cur1, data, s1->name, s1->len) < 0) + if (utf8ncursor(&cur1, um, UTF8_NFDICF, s1->name, s1->len) < 0) return -EINVAL; do { @@ -105,11 +99,10 @@ EXPORT_SYMBOL(utf8_strncasecmp_folded); int utf8_casefold(const struct unicode_map *um, const struct qstr *str, unsigned char *dest, size_t dlen) { - const struct utf8data *data = utf8nfdicf(um->version); struct utf8cursor cur; size_t nlen = 0; - if (utf8ncursor(&cur, data, str->name, str->len) < 0) + if (utf8ncursor(&cur, um, UTF8_NFDICF, str->name, str->len) < 0) return -EINVAL; for (nlen = 0; nlen < dlen; nlen++) { @@ -128,12 +121,11 @@ EXPORT_SYMBOL(utf8_casefold); int utf8_casefold_hash(const struct unicode_map *um, const void *salt, struct qstr *str) { - const struct utf8data *data = utf8nfdicf(um->version); struct utf8cursor cur; int c; unsigned long hash = init_name_hash(salt); - if (utf8ncursor(&cur, data, str->name, str->len) < 0) + if (utf8ncursor(&cur, um, UTF8_NFDICF, str->name, str->len) < 0) return -EINVAL; while ((c = utf8byte(&cur))) { @@ -149,11 +141,10 @@ EXPORT_SYMBOL(utf8_casefold_hash); int utf8_normalize(const struct unicode_map *um, const struct qstr *str, unsigned char *dest, size_t dlen) { - const struct utf8data *data = utf8nfdi(um->version); struct utf8cursor cur; ssize_t nlen = 0; - if (utf8ncursor(&cur, data, str->name, str->len) < 0) + if (utf8ncursor(&cur, um, UTF8_NFDI, str->name, str->len) < 0) return -EINVAL; for (nlen = 0; nlen < dlen; nlen++) { @@ -167,69 +158,59 @@ int utf8_normalize(const struct unicode_map *um, const struct qstr *str, } return -EINVAL; } - EXPORT_SYMBOL(utf8_normalize); -static int utf8_parse_version(const char *version, unsigned int *maj, - unsigned int *min, unsigned int *rev) +static const struct utf8data *find_table_version(const struct utf8data *table, + size_t nr_entries, unsigned int version) { - substring_t args[3]; - char version_string[12]; - static const struct match_token token[] = { - {1, "%d.%d.%d"}, - {0, NULL} - }; - - strncpy(version_string, version, sizeof(version_string)); - - if (match_token(version_string, token, args) != 1) - return -EINVAL; - - if (match_int(&args[0], maj) || match_int(&args[1], min) || - match_int(&args[2], rev)) - return -EINVAL; + size_t i = nr_entries - 1; - return 0; + while (version < table[i].maxage) + i--; + if (version > table[i].maxage) + return NULL; + return &table[i]; } -struct unicode_map *utf8_load(const char *version) +struct unicode_map *utf8_load(unsigned int version) { - struct unicode_map *um = NULL; - int unicode_version; - - if (version) { - unsigned int maj, min, rev; - - if (utf8_parse_version(version, &maj, &min, &rev) < 0) - return ERR_PTR(-EINVAL); - - if (!utf8version_is_supported(maj, min, rev)) - return ERR_PTR(-EINVAL); - - unicode_version = UNICODE_AGE(maj, min, rev); - } else { - unicode_version = utf8version_latest(); - printk(KERN_WARNING"UTF-8 version not specified. " - "Assuming latest supported version (%d.%d.%d).", - (unicode_version >> 16) & 0xff, - (unicode_version >> 8) & 0xff, - (unicode_version & 0xff)); - } + struct unicode_map *um; um = kzalloc(sizeof(struct unicode_map), GFP_KERNEL); if (!um) return ERR_PTR(-ENOMEM); - - um->charset = "UTF-8"; - um->version = unicode_version; - + um->version = version; + + um->tables = symbol_request(utf8_data_table); + if (!um->tables) + goto out_free_um; + + if (!utf8version_is_supported(um, version)) + goto out_symbol_put; + um->ntab[UTF8_NFDI] = find_table_version(um->tables->utf8nfdidata, + um->tables->utf8nfdidata_size, um->version); + if (!um->ntab[UTF8_NFDI]) + goto out_symbol_put; + um->ntab[UTF8_NFDICF] = find_table_version(um->tables->utf8nfdicfdata, + um->tables->utf8nfdicfdata_size, um->version); + if (!um->ntab[UTF8_NFDICF]) + goto out_symbol_put; return um; + +out_symbol_put: + symbol_put(um->tables); +out_free_um: + kfree(um); + return ERR_PTR(-EINVAL); } EXPORT_SYMBOL(utf8_load); void utf8_unload(struct unicode_map *um) { - kfree(um); + if (um) { + symbol_put(utf8_data_table); + kfree(um); + } } EXPORT_SYMBOL(utf8_unload); diff --git a/fs/unicode/utf8-norm.c b/fs/unicode/utf8-norm.c index 1d2d2e5b906a..768f8ab448b8 100644 --- a/fs/unicode/utf8-norm.c +++ b/fs/unicode/utf8-norm.c @@ -6,34 +6,17 @@ #include "utf8n.h" -struct utf8data { - unsigned int maxage; - unsigned int offset; -}; - -#define __INCLUDED_FROM_UTF8NORM_C__ -#include "utf8data.h" -#undef __INCLUDED_FROM_UTF8NORM_C__ - -int utf8version_is_supported(u8 maj, u8 min, u8 rev) +int utf8version_is_supported(const struct unicode_map *um, unsigned int version) { - int i = ARRAY_SIZE(utf8agetab) - 1; - unsigned int sb_utf8version = UNICODE_AGE(maj, min, rev); + int i = um->tables->utf8agetab_size - 1; - while (i >= 0 && utf8agetab[i] != 0) { - if (sb_utf8version == utf8agetab[i]) + while (i >= 0 && um->tables->utf8agetab[i] != 0) { + if (version == um->tables->utf8agetab[i]) return 1; i--; } return 0; } -EXPORT_SYMBOL(utf8version_is_supported); - -int utf8version_latest(void) -{ - return utf8vers; -} -EXPORT_SYMBOL(utf8version_latest); /* * UTF-8 valid ranges. @@ -168,7 +151,7 @@ typedef const unsigned char utf8trie_t; * underlying datatype: unsigned char. * * leaf[0]: The unicode version, stored as a generation number that is - * an index into utf8agetab[]. With this we can filter code + * an index into ->utf8agetab[]. With this we can filter code * points based on the unicode version in which they were * defined. The CCC of a non-defined code point is 0. * leaf[1]: Canonical Combining Class. During normalization, we need @@ -316,21 +299,19 @@ utf8hangul(const char *str, unsigned char *hangul) * is well-formed and corresponds to a known unicode code point. The * shorthand for this will be "is valid UTF-8 unicode". */ -static utf8leaf_t *utf8nlookup(const struct utf8data *data, - unsigned char *hangul, const char *s, size_t len) +static utf8leaf_t *utf8nlookup(const struct unicode_map *um, + enum utf8_normalization n, unsigned char *hangul, const char *s, + size_t len) { - utf8trie_t *trie = NULL; + utf8trie_t *trie = um->tables->utf8data + um->ntab[n]->offset; int offlen; int offset; int mask; int node; - if (!data) - return NULL; if (len == 0) return NULL; - trie = utf8data + data->offset; node = 1; while (node) { offlen = (*trie & OFFLEN) >> OFFLEN_SHIFT; @@ -392,172 +373,29 @@ static utf8leaf_t *utf8nlookup(const struct utf8data *data, * * Forwards to utf8nlookup(). */ -static utf8leaf_t *utf8lookup(const struct utf8data *data, - unsigned char *hangul, const char *s) +static utf8leaf_t *utf8lookup(const struct unicode_map *um, + enum utf8_normalization n, unsigned char *hangul, const char *s) { - return utf8nlookup(data, hangul, s, (size_t)-1); -} - -/* - * Maximum age of any character in s. - * Return -1 if s is not valid UTF-8 unicode. - * Return 0 if only non-assigned code points are used. - */ -int utf8agemax(const struct utf8data *data, const char *s) -{ - utf8leaf_t *leaf; - int age = 0; - int leaf_age; - unsigned char hangul[UTF8HANGULLEAF]; - - if (!data) - return -1; - - while (*s) { - leaf = utf8lookup(data, hangul, s); - if (!leaf) - return -1; - - leaf_age = utf8agetab[LEAF_GEN(leaf)]; - if (leaf_age <= data->maxage && leaf_age > age) - age = leaf_age; - s += utf8clen(s); - } - return age; + return utf8nlookup(um, n, hangul, s, (size_t)-1); } -EXPORT_SYMBOL(utf8agemax); - -/* - * Minimum age of any character in s. - * Return -1 if s is not valid UTF-8 unicode. - * Return 0 if non-assigned code points are used. - */ -int utf8agemin(const struct utf8data *data, const char *s) -{ - utf8leaf_t *leaf; - int age; - int leaf_age; - unsigned char hangul[UTF8HANGULLEAF]; - - if (!data) - return -1; - age = data->maxage; - while (*s) { - leaf = utf8lookup(data, hangul, s); - if (!leaf) - return -1; - leaf_age = utf8agetab[LEAF_GEN(leaf)]; - if (leaf_age <= data->maxage && leaf_age < age) - age = leaf_age; - s += utf8clen(s); - } - return age; -} -EXPORT_SYMBOL(utf8agemin); - -/* - * Maximum age of any character in s, touch at most len bytes. - * Return -1 if s is not valid UTF-8 unicode. - */ -int utf8nagemax(const struct utf8data *data, const char *s, size_t len) -{ - utf8leaf_t *leaf; - int age = 0; - int leaf_age; - unsigned char hangul[UTF8HANGULLEAF]; - - if (!data) - return -1; - - while (len && *s) { - leaf = utf8nlookup(data, hangul, s, len); - if (!leaf) - return -1; - leaf_age = utf8agetab[LEAF_GEN(leaf)]; - if (leaf_age <= data->maxage && leaf_age > age) - age = leaf_age; - len -= utf8clen(s); - s += utf8clen(s); - } - return age; -} -EXPORT_SYMBOL(utf8nagemax); - -/* - * Maximum age of any character in s, touch at most len bytes. - * Return -1 if s is not valid UTF-8 unicode. - */ -int utf8nagemin(const struct utf8data *data, const char *s, size_t len) -{ - utf8leaf_t *leaf; - int leaf_age; - int age; - unsigned char hangul[UTF8HANGULLEAF]; - - if (!data) - return -1; - age = data->maxage; - while (len && *s) { - leaf = utf8nlookup(data, hangul, s, len); - if (!leaf) - return -1; - leaf_age = utf8agetab[LEAF_GEN(leaf)]; - if (leaf_age <= data->maxage && leaf_age < age) - age = leaf_age; - len -= utf8clen(s); - s += utf8clen(s); - } - return age; -} -EXPORT_SYMBOL(utf8nagemin); - -/* - * Length of the normalization of s. - * Return -1 if s is not valid UTF-8 unicode. - * - * A string of Default_Ignorable_Code_Point has length 0. - */ -ssize_t utf8len(const struct utf8data *data, const char *s) -{ - utf8leaf_t *leaf; - size_t ret = 0; - unsigned char hangul[UTF8HANGULLEAF]; - - if (!data) - return -1; - while (*s) { - leaf = utf8lookup(data, hangul, s); - if (!leaf) - return -1; - if (utf8agetab[LEAF_GEN(leaf)] > data->maxage) - ret += utf8clen(s); - else if (LEAF_CCC(leaf) == DECOMPOSE) - ret += strlen(LEAF_STR(leaf)); - else - ret += utf8clen(s); - s += utf8clen(s); - } - return ret; -} -EXPORT_SYMBOL(utf8len); /* * Length of the normalization of s, touch at most len bytes. * Return -1 if s is not valid UTF-8 unicode. */ -ssize_t utf8nlen(const struct utf8data *data, const char *s, size_t len) +ssize_t utf8nlen(const struct unicode_map *um, enum utf8_normalization n, + const char *s, size_t len) { utf8leaf_t *leaf; size_t ret = 0; unsigned char hangul[UTF8HANGULLEAF]; - if (!data) - return -1; while (len && *s) { - leaf = utf8nlookup(data, hangul, s, len); + leaf = utf8nlookup(um, n, hangul, s, len); if (!leaf) return -1; - if (utf8agetab[LEAF_GEN(leaf)] > data->maxage) + if (um->tables->utf8agetab[LEAF_GEN(leaf)] > + um->ntab[n]->maxage) ret += utf8clen(s); else if (LEAF_CCC(leaf) == DECOMPOSE) ret += strlen(LEAF_STR(leaf)); @@ -568,7 +406,6 @@ ssize_t utf8nlen(const struct utf8data *data, const char *s, size_t len) } return ret; } -EXPORT_SYMBOL(utf8nlen); /* * Set up an utf8cursor for use by utf8byte(). @@ -580,14 +417,13 @@ EXPORT_SYMBOL(utf8nlen); * * Returns -1 on error, 0 on success. */ -int utf8ncursor(struct utf8cursor *u8c, const struct utf8data *data, - const char *s, size_t len) +int utf8ncursor(struct utf8cursor *u8c, const struct unicode_map *um, + enum utf8_normalization n, const char *s, size_t len) { - if (!data) - return -1; if (!s) return -1; - u8c->data = data; + u8c->um = um; + u8c->n = n; u8c->s = s; u8c->p = NULL; u8c->ss = NULL; @@ -604,23 +440,6 @@ int utf8ncursor(struct utf8cursor *u8c, const struct utf8data *data, return -1; return 0; } -EXPORT_SYMBOL(utf8ncursor); - -/* - * Set up an utf8cursor for use by utf8byte(). - * - * u8c : pointer to cursor. - * data : const struct utf8data to use for normalization. - * s : NUL-terminated string. - * - * Returns -1 on error, 0 on success. - */ -int utf8cursor(struct utf8cursor *u8c, const struct utf8data *data, - const char *s) -{ - return utf8ncursor(u8c, data, s, (unsigned int)-1); -} -EXPORT_SYMBOL(utf8cursor); /* * Get one byte from the normalized form of the string described by u8c. @@ -678,9 +497,9 @@ int utf8byte(struct utf8cursor *u8c) /* Look up the data for the current character. */ if (u8c->p) { - leaf = utf8lookup(u8c->data, u8c->hangul, u8c->s); + leaf = utf8lookup(u8c->um, u8c->n, u8c->hangul, u8c->s); } else { - leaf = utf8nlookup(u8c->data, u8c->hangul, + leaf = utf8nlookup(u8c->um, u8c->n, u8c->hangul, u8c->s, u8c->len); } @@ -690,7 +509,8 @@ int utf8byte(struct utf8cursor *u8c) ccc = LEAF_CCC(leaf); /* Characters that are too new have CCC 0. */ - if (utf8agetab[LEAF_GEN(leaf)] > u8c->data->maxage) { + if (u8c->um->tables->utf8agetab[LEAF_GEN(leaf)] > + u8c->um->ntab[u8c->n]->maxage) { ccc = STOPPER; } else if (ccc == DECOMPOSE) { u8c->len -= utf8clen(u8c->s); @@ -704,7 +524,7 @@ int utf8byte(struct utf8cursor *u8c) goto ccc_mismatch; } - leaf = utf8lookup(u8c->data, u8c->hangul, u8c->s); + leaf = utf8lookup(u8c->um, u8c->n, u8c->hangul, u8c->s); if (!leaf) return -1; ccc = LEAF_CCC(leaf); @@ -765,28 +585,10 @@ ccc_mismatch: } } } -EXPORT_SYMBOL(utf8byte); - -const struct utf8data *utf8nfdi(unsigned int maxage) -{ - int i = ARRAY_SIZE(utf8nfdidata) - 1; - - while (maxage < utf8nfdidata[i].maxage) - i--; - if (maxage > utf8nfdidata[i].maxage) - return NULL; - return &utf8nfdidata[i]; -} -EXPORT_SYMBOL(utf8nfdi); - -const struct utf8data *utf8nfdicf(unsigned int maxage) -{ - int i = ARRAY_SIZE(utf8nfdicfdata) - 1; - while (maxage < utf8nfdicfdata[i].maxage) - i--; - if (maxage > utf8nfdicfdata[i].maxage) - return NULL; - return &utf8nfdicfdata[i]; -} -EXPORT_SYMBOL(utf8nfdicf); +#ifdef CONFIG_UNICODE_NORMALIZATION_SELFTEST_MODULE +EXPORT_SYMBOL_GPL(utf8version_is_supported); +EXPORT_SYMBOL_GPL(utf8nlen); +EXPORT_SYMBOL_GPL(utf8ncursor); +EXPORT_SYMBOL_GPL(utf8byte); +#endif diff --git a/fs/unicode/utf8-selftest.c b/fs/unicode/utf8-selftest.c index 6fe8af7edccb..eb2bbdd688d7 100644 --- a/fs/unicode/utf8-selftest.c +++ b/fs/unicode/utf8-selftest.c @@ -18,9 +18,7 @@ unsigned int failed_tests; unsigned int total_tests; /* Tests will be based on this version. */ -#define latest_maj 12 -#define latest_min 1 -#define latest_rev 0 +#define UTF8_LATEST UNICODE_AGE(12, 1, 0) #define _test(cond, func, line, fmt, ...) do { \ total_tests++; \ @@ -160,18 +158,22 @@ static const struct { } }; -static void check_utf8_nfdi(void) +static ssize_t utf8len(const struct unicode_map *um, enum utf8_normalization n, + const char *s) +{ + return utf8nlen(um, n, s, (size_t)-1); +} + +static int utf8cursor(struct utf8cursor *u8c, const struct unicode_map *um, + enum utf8_normalization n, const char *s) +{ + return utf8ncursor(u8c, um, n, s, (unsigned int)-1); +} + +static void check_utf8_nfdi(struct unicode_map *um) { int i; struct utf8cursor u8c; - const struct utf8data *data; - - data = utf8nfdi(UNICODE_AGE(latest_maj, latest_min, latest_rev)); - if (!data) { - pr_err("%s: Unable to load utf8-%d.%d.%d. Skipping.\n", - __func__, latest_maj, latest_min, latest_rev); - return; - } for (i = 0; i < ARRAY_SIZE(nfdi_test_data); i++) { int len = strlen(nfdi_test_data[i].str); @@ -179,10 +181,11 @@ static void check_utf8_nfdi(void) int j = 0; unsigned char c; - test((utf8len(data, nfdi_test_data[i].str) == nlen)); - test((utf8nlen(data, nfdi_test_data[i].str, len) == nlen)); + test((utf8len(um, UTF8_NFDI, nfdi_test_data[i].str) == nlen)); + test((utf8nlen(um, UTF8_NFDI, nfdi_test_data[i].str, len) == + nlen)); - if (utf8cursor(&u8c, data, nfdi_test_data[i].str) < 0) + if (utf8cursor(&u8c, um, UTF8_NFDI, nfdi_test_data[i].str) < 0) pr_err("can't create cursor\n"); while ((c = utf8byte(&u8c)) > 0) { @@ -196,18 +199,10 @@ static void check_utf8_nfdi(void) } } -static void check_utf8_nfdicf(void) +static void check_utf8_nfdicf(struct unicode_map *um) { int i; struct utf8cursor u8c; - const struct utf8data *data; - - data = utf8nfdicf(UNICODE_AGE(latest_maj, latest_min, latest_rev)); - if (!data) { - pr_err("%s: Unable to load utf8-%d.%d.%d. Skipping.\n", - __func__, latest_maj, latest_min, latest_rev); - return; - } for (i = 0; i < ARRAY_SIZE(nfdicf_test_data); i++) { int len = strlen(nfdicf_test_data[i].str); @@ -215,10 +210,13 @@ static void check_utf8_nfdicf(void) int j = 0; unsigned char c; - test((utf8len(data, nfdicf_test_data[i].str) == nlen)); - test((utf8nlen(data, nfdicf_test_data[i].str, len) == nlen)); + test((utf8len(um, UTF8_NFDICF, nfdicf_test_data[i].str) == + nlen)); + test((utf8nlen(um, UTF8_NFDICF, nfdicf_test_data[i].str, len) == + nlen)); - if (utf8cursor(&u8c, data, nfdicf_test_data[i].str) < 0) + if (utf8cursor(&u8c, um, UTF8_NFDICF, + nfdicf_test_data[i].str) < 0) pr_err("can't create cursor\n"); while ((c = utf8byte(&u8c)) > 0) { @@ -232,16 +230,9 @@ static void check_utf8_nfdicf(void) } } -static void check_utf8_comparisons(void) +static void check_utf8_comparisons(struct unicode_map *table) { int i; - struct unicode_map *table = utf8_load("12.1.0"); - - if (IS_ERR(table)) { - pr_err("%s: Unable to load utf8 %d.%d.%d. Skipping.\n", - __func__, latest_maj, latest_min, latest_rev); - return; - } for (i = 0; i < ARRAY_SIZE(nfdi_test_data); i++) { const struct qstr s1 = {.name = nfdi_test_data[i].str, @@ -262,42 +253,49 @@ static void check_utf8_comparisons(void) test_f(!utf8_strncasecmp(table, &s1, &s2), "%s %s comparison mismatch\n", s1.name, s2.name); } - - utf8_unload(table); } -static void check_supported_versions(void) +static void check_supported_versions(struct unicode_map *um) { /* Unicode 7.0.0 should be supported. */ - test(utf8version_is_supported(7, 0, 0)); + test(utf8version_is_supported(um, UNICODE_AGE(7, 0, 0))); /* Unicode 9.0.0 should be supported. */ - test(utf8version_is_supported(9, 0, 0)); + test(utf8version_is_supported(um, UNICODE_AGE(9, 0, 0))); /* Unicode 1x.0.0 (the latest version) should be supported. */ - test(utf8version_is_supported(latest_maj, latest_min, latest_rev)); + test(utf8version_is_supported(um, UTF8_LATEST)); /* Next versions don't exist. */ - test(!utf8version_is_supported(13, 0, 0)); - test(!utf8version_is_supported(0, 0, 0)); - test(!utf8version_is_supported(-1, -1, -1)); + test(!utf8version_is_supported(um, UNICODE_AGE(13, 0, 0))); + test(!utf8version_is_supported(um, UNICODE_AGE(0, 0, 0))); + test(!utf8version_is_supported(um, UNICODE_AGE(-1, -1, -1))); } static int __init init_test_ucd(void) { + struct unicode_map *um; + failed_tests = 0; total_tests = 0; - check_supported_versions(); - check_utf8_nfdi(); - check_utf8_nfdicf(); - check_utf8_comparisons(); + um = utf8_load(UTF8_LATEST); + if (IS_ERR(um)) { + pr_err("%s: Unable to load utf8 table.\n", __func__); + return PTR_ERR(um); + } + + check_supported_versions(um); + check_utf8_nfdi(um); + check_utf8_nfdicf(um); + check_utf8_comparisons(um); if (!failed_tests) pr_info("All %u tests passed\n", total_tests); else pr_err("%u out of %u tests failed\n", failed_tests, total_tests); + utf8_unload(um); return 0; } diff --git a/fs/unicode/utf8data.h_shipped b/fs/unicode/utf8data.c_shipped index 76e4f0e1b089..d9b62901aa96 100644 --- a/fs/unicode/utf8data.h_shipped +++ b/fs/unicode/utf8data.c_shipped @@ -1,9 +1,8 @@ /* This file is generated code, do not edit. */ -#ifndef __INCLUDED_FROM_UTF8NORM_C__ -#error Only nls_utf8-norm.c should include this file. -#endif -static const unsigned int utf8vers = 0xc0100; +#include <linux/module.h> +#include <linux/kernel.h> +#include "utf8n.h" static const unsigned int utf8agetab[] = { 0, @@ -4107,3 +4106,18 @@ static const unsigned char utf8data[64256] = { 0x52,0x04,0x00,0x00,0x11,0x04,0x00,0x00,0x02,0x00,0xcf,0x86,0xcf,0x06,0x02,0x00, 0x81,0x80,0xcf,0x86,0x85,0x84,0xcf,0x86,0xcf,0x06,0x02,0x00,0x00,0x00,0x00,0x00 }; + +struct utf8data_table utf8_data_table = { + .utf8agetab = utf8agetab, + .utf8agetab_size = ARRAY_SIZE(utf8agetab), + + .utf8nfdicfdata = utf8nfdicfdata, + .utf8nfdicfdata_size = ARRAY_SIZE(utf8nfdicfdata), + + .utf8nfdidata = utf8nfdidata, + .utf8nfdidata_size = ARRAY_SIZE(utf8nfdidata), + + .utf8data = utf8data, +}; +EXPORT_SYMBOL_GPL(utf8_data_table); +MODULE_LICENSE("GPL v2"); diff --git a/fs/unicode/utf8n.h b/fs/unicode/utf8n.h index 0acd530c2c79..bd00d587747a 100644 --- a/fs/unicode/utf8n.h +++ b/fs/unicode/utf8n.h @@ -11,53 +11,9 @@ #include <linux/export.h> #include <linux/string.h> #include <linux/module.h> +#include <linux/unicode.h> -/* Encoding a unicode version number as a single unsigned int. */ -#define UNICODE_MAJ_SHIFT (16) -#define UNICODE_MIN_SHIFT (8) - -#define UNICODE_AGE(MAJ, MIN, REV) \ - (((unsigned int)(MAJ) << UNICODE_MAJ_SHIFT) | \ - ((unsigned int)(MIN) << UNICODE_MIN_SHIFT) | \ - ((unsigned int)(REV))) - -/* Highest unicode version supported by the data tables. */ -extern int utf8version_is_supported(u8 maj, u8 min, u8 rev); -extern int utf8version_latest(void); - -/* - * Look for the correct const struct utf8data for a unicode version. - * Returns NULL if the version requested is too new. - * - * Two normalization forms are supported: nfdi and nfdicf. - * - * nfdi: - * - Apply unicode normalization form NFD. - * - Remove any Default_Ignorable_Code_Point. - * - * nfdicf: - * - Apply unicode normalization form NFD. - * - Remove any Default_Ignorable_Code_Point. - * - Apply a full casefold (C + F). - */ -extern const struct utf8data *utf8nfdi(unsigned int maxage); -extern const struct utf8data *utf8nfdicf(unsigned int maxage); - -/* - * Determine the maximum age of any unicode character in the string. - * Returns 0 if only unassigned code points are present. - * Returns -1 if the input is not valid UTF-8. - */ -extern int utf8agemax(const struct utf8data *data, const char *s); -extern int utf8nagemax(const struct utf8data *data, const char *s, size_t len); - -/* - * Determine the minimum age of any unicode character in the string. - * Returns 0 if any unassigned code points are present. - * Returns -1 if the input is not valid UTF-8. - */ -extern int utf8agemin(const struct utf8data *data, const char *s); -extern int utf8nagemin(const struct utf8data *data, const char *s, size_t len); +int utf8version_is_supported(const struct unicode_map *um, unsigned int version); /* * Determine the length of the normalized from of the string, @@ -65,8 +21,8 @@ extern int utf8nagemin(const struct utf8data *data, const char *s, size_t len); * Returns 0 if only ignorable code points are present. * Returns -1 if the input is not valid UTF-8. */ -extern ssize_t utf8len(const struct utf8data *data, const char *s); -extern ssize_t utf8nlen(const struct utf8data *data, const char *s, size_t len); +ssize_t utf8nlen(const struct unicode_map *um, enum utf8_normalization n, + const char *s, size_t len); /* Needed in struct utf8cursor below. */ #define UTF8HANGULLEAF (12) @@ -75,7 +31,8 @@ extern ssize_t utf8nlen(const struct utf8data *data, const char *s, size_t len); * Cursor structure used by the normalizer. */ struct utf8cursor { - const struct utf8data *data; + const struct unicode_map *um; + enum utf8_normalization n; const char *s; const char *p; const char *ss; @@ -92,10 +49,8 @@ struct utf8cursor { * Returns 0 on success. * Returns -1 on failure. */ -extern int utf8cursor(struct utf8cursor *u8c, const struct utf8data *data, - const char *s); -extern int utf8ncursor(struct utf8cursor *u8c, const struct utf8data *data, - const char *s, size_t len); +int utf8ncursor(struct utf8cursor *u8c, const struct unicode_map *um, + enum utf8_normalization n, const char *s, size_t len); /* * Get the next byte in the normalization. @@ -105,4 +60,24 @@ extern int utf8ncursor(struct utf8cursor *u8c, const struct utf8data *data, */ extern int utf8byte(struct utf8cursor *u8c); +struct utf8data { + unsigned int maxage; + unsigned int offset; +}; + +struct utf8data_table { + const unsigned int *utf8agetab; + int utf8agetab_size; + + const struct utf8data *utf8nfdicfdata; + int utf8nfdicfdata_size; + + const struct utf8data *utf8nfdidata; + int utf8nfdidata_size; + + const unsigned char *utf8data; +}; + +extern struct utf8data_table utf8_data_table; + #endif /* UTF8NORM_H */ diff --git a/fs/userfaultfd.c b/fs/userfaultfd.c index 22bf14ab2d16..e26b10132d47 100644 --- a/fs/userfaultfd.c +++ b/fs/userfaultfd.c @@ -15,6 +15,7 @@ #include <linux/sched/signal.h> #include <linux/sched/mm.h> #include <linux/mm.h> +#include <linux/mm_inline.h> #include <linux/mmu_notifier.h> #include <linux/poll.h> #include <linux/slab.h> @@ -877,7 +878,7 @@ static int userfaultfd_release(struct inode *inode, struct file *file) new_flags, vma->anon_vma, vma->vm_file, vma->vm_pgoff, vma_policy(vma), - NULL_VM_UFFD_CTX); + NULL_VM_UFFD_CTX, vma_anon_name(vma)); if (prev) vma = prev; else @@ -1436,7 +1437,8 @@ static int userfaultfd_register(struct userfaultfd_ctx *ctx, prev = vma_merge(mm, prev, start, vma_end, new_flags, vma->anon_vma, vma->vm_file, vma->vm_pgoff, vma_policy(vma), - ((struct vm_userfaultfd_ctx){ ctx })); + ((struct vm_userfaultfd_ctx){ ctx }), + vma_anon_name(vma)); if (prev) { vma = prev; goto next; @@ -1613,7 +1615,7 @@ static int userfaultfd_unregister(struct userfaultfd_ctx *ctx, prev = vma_merge(mm, prev, start, vma_end, new_flags, vma->anon_vma, vma->vm_file, vma->vm_pgoff, vma_policy(vma), - NULL_VM_UFFD_CTX); + NULL_VM_UFFD_CTX, vma_anon_name(vma)); if (prev) { vma = prev; goto next; diff --git a/fs/xfs/kmem.c b/fs/xfs/kmem.c index 6f49bf39183c..c557a030acfe 100644 --- a/fs/xfs/kmem.c +++ b/fs/xfs/kmem.c @@ -4,7 +4,6 @@ * All Rights Reserved. */ #include "xfs.h" -#include <linux/backing-dev.h> #include "xfs_message.h" #include "xfs_trace.h" @@ -26,6 +25,6 @@ kmem_alloc(size_t size, xfs_km_flags_t flags) "%s(%u) possible memory allocation deadlock size %u in %s (mode:0x%x)", current->comm, current->pid, (unsigned int)size, __func__, lflags); - congestion_wait(BLK_RW_ASYNC, HZ/50); + memalloc_retry_wait(lflags); } while (1); } diff --git a/fs/xfs/libxfs/xfs_fs.h b/fs/xfs/libxfs/xfs_fs.h index c43877c8a279..505533c43a92 100644 --- a/fs/xfs/libxfs/xfs_fs.h +++ b/fs/xfs/libxfs/xfs_fs.h @@ -93,21 +93,6 @@ struct getbmapx { #define XFS_FMR_OWN_DEFECTIVE FMR_OWNER('X', 8) /* bad blocks */ /* - * Structure for XFS_IOC_FSSETDM. - * For use by backup and restore programs to set the XFS on-disk inode - * fields di_dmevmask and di_dmstate. These must be set to exactly and - * only values previously obtained via xfs_bulkstat! (Specifically the - * struct xfs_bstat fields bs_dmevmask and bs_dmstate.) - */ -#ifndef HAVE_FSDMIDATA -struct fsdmidata { - __u32 fsd_dmevmask; /* corresponds to di_dmevmask */ - __u16 fsd_padding; - __u16 fsd_dmstate; /* corresponds to di_dmstate */ -}; -#endif - -/* * File segment locking set data type for 64 bit access. * Also used for all the RESV/FREE interfaces. */ @@ -562,16 +547,10 @@ typedef struct xfs_fsop_handlereq { /* * Compound structures for passing args through Handle Request interfaces - * xfs_fssetdm_by_handle, xfs_attrlist_by_handle, xfs_attrmulti_by_handle - * - ioctls: XFS_IOC_FSSETDM_BY_HANDLE, XFS_IOC_ATTRLIST_BY_HANDLE, and - * XFS_IOC_ATTRMULTI_BY_HANDLE + * xfs_attrlist_by_handle, xfs_attrmulti_by_handle + * - ioctls: XFS_IOC_ATTRLIST_BY_HANDLE, and XFS_IOC_ATTRMULTI_BY_HANDLE */ -typedef struct xfs_fsop_setdm_handlereq { - struct xfs_fsop_handlereq hreq; /* handle information */ - struct fsdmidata __user *data; /* DMAPI data */ -} xfs_fsop_setdm_handlereq_t; - /* * Flags passed in xfs_attr_multiop.am_flags for the attr ioctl interface. * @@ -781,15 +760,15 @@ struct xfs_scrub_metadata { * For 'documentation' purposed more than anything else, * the "cmd #" field reflects the IRIX fcntl number. */ -#define XFS_IOC_ALLOCSP _IOW ('X', 10, struct xfs_flock64) -#define XFS_IOC_FREESP _IOW ('X', 11, struct xfs_flock64) +/* XFS_IOC_ALLOCSP ------- deprecated 10 */ +/* XFS_IOC_FREESP -------- deprecated 11 */ #define XFS_IOC_DIOINFO _IOR ('X', 30, struct dioattr) #define XFS_IOC_FSGETXATTR FS_IOC_FSGETXATTR #define XFS_IOC_FSSETXATTR FS_IOC_FSSETXATTR -#define XFS_IOC_ALLOCSP64 _IOW ('X', 36, struct xfs_flock64) -#define XFS_IOC_FREESP64 _IOW ('X', 37, struct xfs_flock64) +/* XFS_IOC_ALLOCSP64 ----- deprecated 36 */ +/* XFS_IOC_FREESP64 ------ deprecated 37 */ #define XFS_IOC_GETBMAP _IOWR('X', 38, struct getbmap) -#define XFS_IOC_FSSETDM _IOW ('X', 39, struct fsdmidata) +/* XFS_IOC_FSSETDM ------- deprecated 39 */ #define XFS_IOC_RESVSP _IOW ('X', 40, struct xfs_flock64) #define XFS_IOC_UNRESVSP _IOW ('X', 41, struct xfs_flock64) #define XFS_IOC_RESVSP64 _IOW ('X', 42, struct xfs_flock64) @@ -831,7 +810,7 @@ struct xfs_scrub_metadata { #define XFS_IOC_FREEZE _IOWR('X', 119, int) /* aka FIFREEZE */ #define XFS_IOC_THAW _IOWR('X', 120, int) /* aka FITHAW */ -#define XFS_IOC_FSSETDM_BY_HANDLE _IOW ('X', 121, struct xfs_fsop_setdm_handlereq) +/* XFS_IOC_FSSETDM_BY_HANDLE -- deprecated 121 */ #define XFS_IOC_ATTRLIST_BY_HANDLE _IOW ('X', 122, struct xfs_fsop_attrlist_handlereq) #define XFS_IOC_ATTRMULTI_BY_HANDLE _IOW ('X', 123, struct xfs_fsop_attrmulti_handlereq) #define XFS_IOC_FSGEOMETRY_V4 _IOR ('X', 124, struct xfs_fsop_geom_v4) diff --git a/fs/xfs/scrub/agheader.c b/fs/xfs/scrub/agheader.c index bed798792226..90aebfe9dc5f 100644 --- a/fs/xfs/scrub/agheader.c +++ b/fs/xfs/scrub/agheader.c @@ -281,7 +281,7 @@ xchk_superblock( features_mask = cpu_to_be32(XFS_SB_VERSION2_ATTR2BIT); if ((sb->sb_features2 & features_mask) != (cpu_to_be32(mp->m_sb.sb_features2) & features_mask)) - xchk_block_set_corrupt(sc, bp); + xchk_block_set_preen(sc, bp); if (!xfs_has_crc(mp)) { /* all v5 fields must be zero */ @@ -290,39 +290,38 @@ xchk_superblock( offsetof(struct xfs_dsb, sb_features_compat))) xchk_block_set_corrupt(sc, bp); } else { - /* Check compat flags; all are set at mkfs time. */ - features_mask = cpu_to_be32(XFS_SB_FEAT_COMPAT_UNKNOWN); - if ((sb->sb_features_compat & features_mask) != - (cpu_to_be32(mp->m_sb.sb_features_compat) & features_mask)) + /* compat features must match */ + if (sb->sb_features_compat != + cpu_to_be32(mp->m_sb.sb_features_compat)) xchk_block_set_corrupt(sc, bp); - /* Check ro compat flags; all are set at mkfs time. */ - features_mask = cpu_to_be32(XFS_SB_FEAT_RO_COMPAT_UNKNOWN | - XFS_SB_FEAT_RO_COMPAT_FINOBT | - XFS_SB_FEAT_RO_COMPAT_RMAPBT | - XFS_SB_FEAT_RO_COMPAT_REFLINK); - if ((sb->sb_features_ro_compat & features_mask) != - (cpu_to_be32(mp->m_sb.sb_features_ro_compat) & - features_mask)) + /* ro compat features must match */ + if (sb->sb_features_ro_compat != + cpu_to_be32(mp->m_sb.sb_features_ro_compat)) xchk_block_set_corrupt(sc, bp); - /* Check incompat flags; all are set at mkfs time. */ - features_mask = cpu_to_be32(XFS_SB_FEAT_INCOMPAT_UNKNOWN | - XFS_SB_FEAT_INCOMPAT_FTYPE | - XFS_SB_FEAT_INCOMPAT_SPINODES | - XFS_SB_FEAT_INCOMPAT_META_UUID); - if ((sb->sb_features_incompat & features_mask) != - (cpu_to_be32(mp->m_sb.sb_features_incompat) & - features_mask)) - xchk_block_set_corrupt(sc, bp); + /* + * NEEDSREPAIR is ignored on a secondary super, so we should + * clear it when we find it, though it's not a corruption. + */ + features_mask = cpu_to_be32(XFS_SB_FEAT_INCOMPAT_NEEDSREPAIR); + if ((cpu_to_be32(mp->m_sb.sb_features_incompat) ^ + sb->sb_features_incompat) & features_mask) + xchk_block_set_preen(sc, bp); - /* Check log incompat flags; all are set at mkfs time. */ - features_mask = cpu_to_be32(XFS_SB_FEAT_INCOMPAT_LOG_UNKNOWN); - if ((sb->sb_features_log_incompat & features_mask) != - (cpu_to_be32(mp->m_sb.sb_features_log_incompat) & - features_mask)) + /* all other incompat features must match */ + if ((cpu_to_be32(mp->m_sb.sb_features_incompat) ^ + sb->sb_features_incompat) & ~features_mask) xchk_block_set_corrupt(sc, bp); + /* + * log incompat features protect newer log record types from + * older log recovery code. Log recovery doesn't check the + * secondary supers, so we can clear these if needed. + */ + if (sb->sb_features_log_incompat) + xchk_block_set_preen(sc, bp); + /* Don't care about sb_crc */ if (sb->sb_spino_align != cpu_to_be32(mp->m_sb.sb_spino_align)) diff --git a/fs/xfs/scrub/agheader_repair.c b/fs/xfs/scrub/agheader_repair.c index d7bfed52f4cd..6da7f2ca77de 100644 --- a/fs/xfs/scrub/agheader_repair.c +++ b/fs/xfs/scrub/agheader_repair.c @@ -52,6 +52,18 @@ xrep_superblock( xfs_buf_zero(bp, 0, BBTOB(bp->b_length)); xfs_sb_to_disk(bp->b_addr, &mp->m_sb); + /* + * Don't write out a secondary super with NEEDSREPAIR or log incompat + * features set, since both are ignored when set on a secondary. + */ + if (xfs_has_crc(mp)) { + struct xfs_dsb *sb = bp->b_addr; + + sb->sb_features_incompat &= + ~cpu_to_be32(XFS_SB_FEAT_INCOMPAT_NEEDSREPAIR); + sb->sb_features_log_incompat = 0; + } + /* Write this to disk. */ xfs_trans_buf_set_type(sc->tp, bp, XFS_BLFT_SB_BUF); xfs_trans_log_buf(sc->tp, bp, 0, BBTOB(bp->b_length) - 1); diff --git a/fs/xfs/xfs_bmap_util.c b/fs/xfs/xfs_bmap_util.c index 797ea0c8b14e..d4a387d3d0ce 100644 --- a/fs/xfs/xfs_bmap_util.c +++ b/fs/xfs/xfs_bmap_util.c @@ -771,8 +771,7 @@ int xfs_alloc_file_space( struct xfs_inode *ip, xfs_off_t offset, - xfs_off_t len, - int alloc_type) + xfs_off_t len) { xfs_mount_t *mp = ip->i_mount; xfs_off_t count; @@ -865,8 +864,8 @@ xfs_alloc_file_space( goto error; error = xfs_bmapi_write(tp, ip, startoffset_fsb, - allocatesize_fsb, alloc_type, 0, imapp, - &nimaps); + allocatesize_fsb, XFS_BMAPI_PREALLOC, 0, imapp, + &nimaps); if (error) goto error; diff --git a/fs/xfs/xfs_bmap_util.h b/fs/xfs/xfs_bmap_util.h index 9f993168b55b..24b37d211f1d 100644 --- a/fs/xfs/xfs_bmap_util.h +++ b/fs/xfs/xfs_bmap_util.h @@ -54,7 +54,7 @@ int xfs_bmap_last_extent(struct xfs_trans *tp, struct xfs_inode *ip, /* preallocation and hole punch interface */ int xfs_alloc_file_space(struct xfs_inode *ip, xfs_off_t offset, - xfs_off_t len, int alloc_type); + xfs_off_t len); int xfs_free_file_space(struct xfs_inode *ip, xfs_off_t offset, xfs_off_t len); int xfs_collapse_file_space(struct xfs_inode *, xfs_off_t offset, diff --git a/fs/xfs/xfs_buf.c b/fs/xfs/xfs_buf.c index bbb0fbd34e64..b45e0d50a405 100644 --- a/fs/xfs/xfs_buf.c +++ b/fs/xfs/xfs_buf.c @@ -394,7 +394,7 @@ xfs_buf_alloc_pages( } XFS_STATS_INC(bp->b_mount, xb_page_retries); - congestion_wait(BLK_RW_ASYNC, HZ / 50); + memalloc_retry_wait(gfp_mask); } return 0; } diff --git a/fs/xfs/xfs_dir2_readdir.c b/fs/xfs/xfs_dir2_readdir.c index 8310005af00f..a7174a5b3203 100644 --- a/fs/xfs/xfs_dir2_readdir.c +++ b/fs/xfs/xfs_dir2_readdir.c @@ -138,7 +138,8 @@ xfs_dir2_sf_getdents( STATIC int xfs_dir2_block_getdents( struct xfs_da_args *args, - struct dir_context *ctx) + struct dir_context *ctx, + unsigned int *lock_mode) { struct xfs_inode *dp = args->dp; /* incore directory inode */ struct xfs_buf *bp; /* buffer for block */ @@ -146,7 +147,6 @@ xfs_dir2_block_getdents( int wantoff; /* starting block offset */ xfs_off_t cook; struct xfs_da_geometry *geo = args->geo; - int lock_mode; unsigned int offset, next_offset; unsigned int end; @@ -156,12 +156,13 @@ xfs_dir2_block_getdents( if (xfs_dir2_dataptr_to_db(geo, ctx->pos) > geo->datablk) return 0; - lock_mode = xfs_ilock_data_map_shared(dp); error = xfs_dir3_block_read(args->trans, dp, &bp); - xfs_iunlock(dp, lock_mode); if (error) return error; + xfs_iunlock(dp, *lock_mode); + *lock_mode = 0; + /* * Extract the byte offset we start at from the seek pointer. * We'll skip entries before this. @@ -344,7 +345,8 @@ STATIC int xfs_dir2_leaf_getdents( struct xfs_da_args *args, struct dir_context *ctx, - size_t bufsize) + size_t bufsize, + unsigned int *lock_mode) { struct xfs_inode *dp = args->dp; struct xfs_mount *mp = dp->i_mount; @@ -356,7 +358,6 @@ xfs_dir2_leaf_getdents( xfs_dir2_off_t curoff; /* current overall offset */ int length; /* temporary length value */ int byteoff; /* offset in current block */ - int lock_mode; unsigned int offset = 0; int error = 0; /* error return value */ @@ -390,13 +391,16 @@ xfs_dir2_leaf_getdents( bp = NULL; } - lock_mode = xfs_ilock_data_map_shared(dp); + if (*lock_mode == 0) + *lock_mode = xfs_ilock_data_map_shared(dp); error = xfs_dir2_leaf_readbuf(args, bufsize, &curoff, &rablk, &bp); - xfs_iunlock(dp, lock_mode); if (error || !bp) break; + xfs_iunlock(dp, *lock_mode); + *lock_mode = 0; + xfs_dir3_data_check(dp, bp); /* * Find our position in the block. @@ -496,7 +500,7 @@ xfs_dir2_leaf_getdents( * * If supplied, the transaction collects locked dir buffers to avoid * nested buffer deadlocks. This function does not dirty the - * transaction. The caller should ensure that the inode is locked + * transaction. The caller must hold the IOLOCK (shared or exclusive) * before calling this function. */ int @@ -507,8 +511,9 @@ xfs_readdir( size_t bufsize) { struct xfs_da_args args = { NULL }; - int rval; - int v; + unsigned int lock_mode; + int isblock; + int error; trace_xfs_readdir(dp); @@ -516,6 +521,7 @@ xfs_readdir( return -EIO; ASSERT(S_ISDIR(VFS_I(dp)->i_mode)); + ASSERT(xfs_isilocked(dp, XFS_IOLOCK_SHARED | XFS_IOLOCK_EXCL)); XFS_STATS_INC(dp->i_mount, xs_dir_getdents); args.dp = dp; @@ -523,13 +529,22 @@ xfs_readdir( args.trans = tp; if (dp->i_df.if_format == XFS_DINODE_FMT_LOCAL) - rval = xfs_dir2_sf_getdents(&args, ctx); - else if ((rval = xfs_dir2_isblock(&args, &v))) - ; - else if (v) - rval = xfs_dir2_block_getdents(&args, ctx); - else - rval = xfs_dir2_leaf_getdents(&args, ctx, bufsize); + return xfs_dir2_sf_getdents(&args, ctx); - return rval; + lock_mode = xfs_ilock_data_map_shared(dp); + error = xfs_dir2_isblock(&args, &isblock); + if (error) + goto out_unlock; + + if (isblock) { + error = xfs_dir2_block_getdents(&args, ctx, &lock_mode); + goto out_unlock; + } + + error = xfs_dir2_leaf_getdents(&args, ctx, bufsize, &lock_mode); + +out_unlock: + if (lock_mode) + xfs_iunlock(dp, lock_mode); + return error; } diff --git a/fs/xfs/xfs_file.c b/fs/xfs/xfs_file.c index 8d4c5ca261bd..22ad207bedf4 100644 --- a/fs/xfs/xfs_file.c +++ b/fs/xfs/xfs_file.c @@ -1051,8 +1051,7 @@ xfs_file_fallocate( } if (!xfs_is_always_cow_inode(ip)) { - error = xfs_alloc_file_space(ip, offset, len, - XFS_BMAPI_PREALLOC); + error = xfs_alloc_file_space(ip, offset, len); if (error) goto out_unlock; } diff --git a/fs/xfs/xfs_icache.c b/fs/xfs/xfs_icache.c index 2e718728986f..9644f938990c 100644 --- a/fs/xfs/xfs_icache.c +++ b/fs/xfs/xfs_icache.c @@ -1854,28 +1854,20 @@ xfs_inodegc_worker( } /* - * Force all currently queued inode inactivation work to run immediately, and - * wait for the work to finish. Two pass - queue all the work first pass, wait - * for it in a second pass. + * Force all currently queued inode inactivation work to run immediately and + * wait for the work to finish. */ void xfs_inodegc_flush( struct xfs_mount *mp) { - struct xfs_inodegc *gc; - int cpu; - if (!xfs_is_inodegc_enabled(mp)) return; trace_xfs_inodegc_flush(mp, __return_address); xfs_inodegc_queue_all(mp); - - for_each_online_cpu(cpu) { - gc = per_cpu_ptr(mp->m_inodegc, cpu); - flush_work(&gc->work); - } + flush_workqueue(mp->m_inodegc_wq); } /* @@ -1886,18 +1878,12 @@ void xfs_inodegc_stop( struct xfs_mount *mp) { - struct xfs_inodegc *gc; - int cpu; - if (!xfs_clear_inodegc_enabled(mp)) return; xfs_inodegc_queue_all(mp); + drain_workqueue(mp->m_inodegc_wq); - for_each_online_cpu(cpu) { - gc = per_cpu_ptr(mp->m_inodegc, cpu); - cancel_work_sync(&gc->work); - } trace_xfs_inodegc_stop(mp, __return_address); } diff --git a/fs/xfs/xfs_ioctl.c b/fs/xfs/xfs_ioctl.c index 8ea47a9d5aad..03a6198c97f6 100644 --- a/fs/xfs/xfs_ioctl.c +++ b/fs/xfs/xfs_ioctl.c @@ -627,87 +627,6 @@ xfs_attrmulti_by_handle( return error; } -int -xfs_ioc_space( - struct file *filp, - xfs_flock64_t *bf) -{ - struct inode *inode = file_inode(filp); - struct xfs_inode *ip = XFS_I(inode); - struct iattr iattr; - enum xfs_prealloc_flags flags = XFS_PREALLOC_CLEAR; - uint iolock = XFS_IOLOCK_EXCL | XFS_MMAPLOCK_EXCL; - int error; - - if (inode->i_flags & (S_IMMUTABLE|S_APPEND)) - return -EPERM; - - if (!(filp->f_mode & FMODE_WRITE)) - return -EBADF; - - if (!S_ISREG(inode->i_mode)) - return -EINVAL; - - if (xfs_is_always_cow_inode(ip)) - return -EOPNOTSUPP; - - if (filp->f_flags & O_DSYNC) - flags |= XFS_PREALLOC_SYNC; - if (filp->f_mode & FMODE_NOCMTIME) - flags |= XFS_PREALLOC_INVISIBLE; - - error = mnt_want_write_file(filp); - if (error) - return error; - - xfs_ilock(ip, iolock); - error = xfs_break_layouts(inode, &iolock, BREAK_UNMAP); - if (error) - goto out_unlock; - inode_dio_wait(inode); - - switch (bf->l_whence) { - case 0: /*SEEK_SET*/ - break; - case 1: /*SEEK_CUR*/ - bf->l_start += filp->f_pos; - break; - case 2: /*SEEK_END*/ - bf->l_start += XFS_ISIZE(ip); - break; - default: - error = -EINVAL; - goto out_unlock; - } - - if (bf->l_start < 0 || bf->l_start > inode->i_sb->s_maxbytes) { - error = -EINVAL; - goto out_unlock; - } - - if (bf->l_start > XFS_ISIZE(ip)) { - error = xfs_alloc_file_space(ip, XFS_ISIZE(ip), - bf->l_start - XFS_ISIZE(ip), - XFS_BMAPI_PREALLOC); - if (error) - goto out_unlock; - } - - iattr.ia_valid = ATTR_SIZE; - iattr.ia_size = bf->l_start; - error = xfs_vn_setattr_size(file_mnt_user_ns(filp), file_dentry(filp), - &iattr); - if (error) - goto out_unlock; - - error = xfs_update_prealloc_flags(ip, flags); - -out_unlock: - xfs_iunlock(ip, iolock); - mnt_drop_write_file(filp); - return error; -} - /* Return 0 on success or positive error */ int xfs_fsbulkstat_one_fmt( @@ -1936,6 +1855,15 @@ xfs_fs_eofblocks_from_user( } /* + * These long-unused ioctls were removed from the official ioctl API in 5.17, + * but retain these definitions so that we can log warnings about them. + */ +#define XFS_IOC_ALLOCSP _IOW ('X', 10, struct xfs_flock64) +#define XFS_IOC_FREESP _IOW ('X', 11, struct xfs_flock64) +#define XFS_IOC_ALLOCSP64 _IOW ('X', 36, struct xfs_flock64) +#define XFS_IOC_FREESP64 _IOW ('X', 37, struct xfs_flock64) + +/* * Note: some of the ioctl's return positive numbers as a * byte count indicating success, such as readlink_by_handle. * So we don't "sign flip" like most other routines. This means @@ -1965,13 +1893,11 @@ xfs_file_ioctl( case XFS_IOC_ALLOCSP: case XFS_IOC_FREESP: case XFS_IOC_ALLOCSP64: - case XFS_IOC_FREESP64: { - xfs_flock64_t bf; - - if (copy_from_user(&bf, arg, sizeof(bf))) - return -EFAULT; - return xfs_ioc_space(filp, &bf); - } + case XFS_IOC_FREESP64: + xfs_warn_once(mp, + "%s should use fallocate; XFS_IOC_{ALLOC,FREE}SP ioctl unsupported", + current->comm); + return -ENOTTY; case XFS_IOC_DIOINFO: { struct xfs_buftarg *target = xfs_inode_buftarg(ip); struct dioattr da; diff --git a/fs/xfs/xfs_ioctl.h b/fs/xfs/xfs_ioctl.h index 845d3bcab74b..d4abba2c13c1 100644 --- a/fs/xfs/xfs_ioctl.h +++ b/fs/xfs/xfs_ioctl.h @@ -10,12 +10,6 @@ struct xfs_bstat; struct xfs_ibulk; struct xfs_inogrp; - -extern int -xfs_ioc_space( - struct file *filp, - xfs_flock64_t *bf); - int xfs_ioc_swapext( xfs_swapext_t *sxp); diff --git a/fs/xfs/xfs_ioctl32.c b/fs/xfs/xfs_ioctl32.c index 8783af203cfc..004ed2a251e8 100644 --- a/fs/xfs/xfs_ioctl32.c +++ b/fs/xfs/xfs_ioctl32.c @@ -28,22 +28,6 @@ #ifdef BROKEN_X86_ALIGNMENT STATIC int -xfs_compat_flock64_copyin( - xfs_flock64_t *bf, - compat_xfs_flock64_t __user *arg32) -{ - if (get_user(bf->l_type, &arg32->l_type) || - get_user(bf->l_whence, &arg32->l_whence) || - get_user(bf->l_start, &arg32->l_start) || - get_user(bf->l_len, &arg32->l_len) || - get_user(bf->l_sysid, &arg32->l_sysid) || - get_user(bf->l_pid, &arg32->l_pid) || - copy_from_user(bf->l_pad, &arg32->l_pad, 4*sizeof(u32))) - return -EFAULT; - return 0; -} - -STATIC int xfs_compat_ioc_fsgeometry_v1( struct xfs_mount *mp, compat_xfs_fsop_geom_v1_t __user *arg32) @@ -445,17 +429,6 @@ xfs_file_compat_ioctl( switch (cmd) { #if defined(BROKEN_X86_ALIGNMENT) - case XFS_IOC_ALLOCSP_32: - case XFS_IOC_FREESP_32: - case XFS_IOC_ALLOCSP64_32: - case XFS_IOC_FREESP64_32: { - struct xfs_flock64 bf; - - if (xfs_compat_flock64_copyin(&bf, arg)) - return -EFAULT; - cmd = _NATIVE_IOC(cmd, struct xfs_flock64); - return xfs_ioc_space(filp, &bf); - } case XFS_IOC_FSGEOMETRY_V1_32: return xfs_compat_ioc_fsgeometry_v1(ip->i_mount, arg); case XFS_IOC_FSGROWFSDATA_32: { diff --git a/fs/xfs/xfs_ioctl32.h b/fs/xfs/xfs_ioctl32.h index 9929482bf358..c14852362fce 100644 --- a/fs/xfs/xfs_ioctl32.h +++ b/fs/xfs/xfs_ioctl32.h @@ -142,28 +142,6 @@ typedef struct compat_xfs_fsop_attrmulti_handlereq { _IOW('X', 123, struct compat_xfs_fsop_attrmulti_handlereq) #ifdef BROKEN_X86_ALIGNMENT -/* on ia32 l_start is on a 32-bit boundary */ -typedef struct compat_xfs_flock64 { - __s16 l_type; - __s16 l_whence; - __s64 l_start __attribute__((packed)); - /* len == 0 means until end of file */ - __s64 l_len __attribute__((packed)); - __s32 l_sysid; - __u32 l_pid; - __s32 l_pad[4]; /* reserve area */ -} compat_xfs_flock64_t; - -#define XFS_IOC_ALLOCSP_32 _IOW('X', 10, struct compat_xfs_flock64) -#define XFS_IOC_FREESP_32 _IOW('X', 11, struct compat_xfs_flock64) -#define XFS_IOC_ALLOCSP64_32 _IOW('X', 36, struct compat_xfs_flock64) -#define XFS_IOC_FREESP64_32 _IOW('X', 37, struct compat_xfs_flock64) -#define XFS_IOC_RESVSP_32 _IOW('X', 40, struct compat_xfs_flock64) -#define XFS_IOC_UNRESVSP_32 _IOW('X', 41, struct compat_xfs_flock64) -#define XFS_IOC_RESVSP64_32 _IOW('X', 42, struct compat_xfs_flock64) -#define XFS_IOC_UNRESVSP64_32 _IOW('X', 43, struct compat_xfs_flock64) -#define XFS_IOC_ZERO_RANGE_32 _IOW('X', 57, struct compat_xfs_flock64) - typedef struct compat_xfs_fsop_geom_v1 { __u32 blocksize; /* filesystem (data) block size */ __u32 rtextsize; /* realtime extent size */ |