From 1e8f44f159b31fe31ad2f40f96575b6ad6df2fe9 Mon Sep 17 00:00:00 2001 From: Al Viro Date: Wed, 11 Mar 2020 17:22:19 -0400 Subject: do_tmpfile(): don't mess with finish_open() use vfs_open() instead Signed-off-by: Al Viro --- fs/namei.c | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) diff --git a/fs/namei.c b/fs/namei.c index 78443a85480a..a3b3ca62ef5c 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -3325,10 +3325,8 @@ static int do_tmpfile(struct nameidata *nd, unsigned flags, audit_inode(nd->name, child, 0); /* Don't check for other permissions, the inode was just created */ error = may_open(&path, 0, op->open_flag); - if (error) - goto out2; - file->f_path.mnt = path.mnt; - error = finish_open(file, child, NULL); + if (!error) + error = vfs_open(&path, file); out2: mnt_drop_write(path.mnt); out: -- cgit From 26ddb45e13a3e09f5f5517a3c9d6510208181516 Mon Sep 17 00:00:00 2001 From: "Steven Rostedt (VMware)" Date: Wed, 9 Dec 2020 17:09:28 -0500 Subject: fs/namei.c: Remove unlikely of status being -ECHILD in lookup_fast() Running my yearly branch profiling code, it detected a 100% wrong branch condition in name.c for lookup_fast(). The code in question has: status = d_revalidate(dentry, nd->flags); if (likely(status > 0)) return dentry; if (unlazy_child(nd, dentry, seq)) return ERR_PTR(-ECHILD); if (unlikely(status == -ECHILD)) /* we'd been told to redo it in non-rcu mode */ status = d_revalidate(dentry, nd->flags); If the status of the d_revalidate() is greater than zero, then the function finishes. Otherwise, if it is an "unlazy_child" it returns with -ECHILD. After the above two checks, the status is compared to -ECHILD, as that is what is returned if the original d_revalidate() needed to be done in a non-rcu mode. Especially this path is called in a condition of: if (nd->flags & LOOKUP_RCU) { And most of the d_revalidate() functions have: if (flags & LOOKUP_RCU) return -ECHILD; It appears that that is the only case that this if statement is triggered on two of my machines, running in production. As it is dependent on what filesystem mix is configured in the running kernel, simply remove the unlikely() from the if statement. Signed-off-by: Steven Rostedt (VMware) Signed-off-by: Al Viro --- fs/namei.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/fs/namei.c b/fs/namei.c index a3b3ca62ef5c..3345a9f38ccb 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -1495,7 +1495,7 @@ static struct dentry *lookup_fast(struct nameidata *nd, return dentry; if (unlazy_child(nd, dentry, seq)) return ERR_PTR(-ECHILD); - if (unlikely(status == -ECHILD)) + if (status == -ECHILD) /* we'd been told to redo it in non-rcu mode */ status = d_revalidate(dentry, nd->flags); } else { -- cgit From e36cffed20a324e116f329a94061ae30dd26fb51 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Thu, 17 Dec 2020 09:19:08 -0700 Subject: fs: make unlazy_walk() error handling consistent Most callers check for non-zero return, and assume it's -ECHILD (which it always will be). One caller uses the actual error return. Clean this up and make it fully consistent, by having unlazy_walk() return a bool instead. Rename it to try_to_unlazy() and return true on success, and failure on error. That's easier to read. No functional changes in this patch. Cc: Al Viro Signed-off-by: Jens Axboe Signed-off-by: Al Viro --- fs/namei.c | 43 +++++++++++++++++-------------------------- 1 file changed, 17 insertions(+), 26 deletions(-) diff --git a/fs/namei.c b/fs/namei.c index 3345a9f38ccb..21fd06753504 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -669,17 +669,17 @@ static bool legitimize_root(struct nameidata *nd) */ /** - * unlazy_walk - try to switch to ref-walk mode. + * try_to_unlazy - try to switch to ref-walk mode. * @nd: nameidata pathwalk data - * Returns: 0 on success, -ECHILD on failure + * Returns: true on success, false on failure * - * unlazy_walk attempts to legitimize the current nd->path and nd->root + * try_to_unlazy attempts to legitimize the current nd->path and nd->root * for ref-walk mode. * Must be called from rcu-walk context. - * Nothing should touch nameidata between unlazy_walk() failure and + * Nothing should touch nameidata between try_to_unlazy() failure and * terminate_walk(). */ -static int unlazy_walk(struct nameidata *nd) +static bool try_to_unlazy(struct nameidata *nd) { struct dentry *parent = nd->path.dentry; @@ -694,14 +694,14 @@ static int unlazy_walk(struct nameidata *nd) goto out; rcu_read_unlock(); BUG_ON(nd->inode != parent->d_inode); - return 0; + return true; out1: nd->path.mnt = NULL; nd->path.dentry = NULL; out: rcu_read_unlock(); - return -ECHILD; + return false; } /** @@ -792,7 +792,7 @@ static int complete_walk(struct nameidata *nd) */ if (!(nd->flags & (LOOKUP_ROOT | LOOKUP_IS_SCOPED))) nd->root.mnt = NULL; - if (unlikely(unlazy_walk(nd))) + if (!try_to_unlazy(nd)) return -ECHILD; } @@ -1466,7 +1466,7 @@ static struct dentry *lookup_fast(struct nameidata *nd, unsigned seq; dentry = __d_lookup_rcu(parent, &nd->last, &seq); if (unlikely(!dentry)) { - if (unlazy_walk(nd)) + if (!try_to_unlazy(nd)) return ERR_PTR(-ECHILD); return NULL; } @@ -1567,10 +1567,8 @@ static inline int may_lookup(struct nameidata *nd) { if (nd->flags & LOOKUP_RCU) { int err = inode_permission(nd->inode, MAY_EXEC|MAY_NOT_BLOCK); - if (err != -ECHILD) + if (err != -ECHILD || !try_to_unlazy(nd)) return err; - if (unlazy_walk(nd)) - return -ECHILD; } return inode_permission(nd->inode, MAY_EXEC); } @@ -1592,7 +1590,7 @@ static int reserve_stack(struct nameidata *nd, struct path *link, unsigned seq) // unlazy even if we fail to grab the link - cleanup needs it bool grabbed_link = legitimize_path(nd, link, seq); - if (unlazy_walk(nd) != 0 || !grabbed_link) + if (!try_to_unlazy(nd) != 0 || !grabbed_link) return -ECHILD; if (nd_alloc_stack(nd)) @@ -1634,7 +1632,7 @@ static const char *pick_link(struct nameidata *nd, struct path *link, touch_atime(&last->link); cond_resched(); } else if (atime_needs_update(&last->link, inode)) { - if (unlikely(unlazy_walk(nd))) + if (!try_to_unlazy(nd)) return ERR_PTR(-ECHILD); touch_atime(&last->link); } @@ -1651,11 +1649,8 @@ static const char *pick_link(struct nameidata *nd, struct path *link, get = inode->i_op->get_link; if (nd->flags & LOOKUP_RCU) { res = get(NULL, inode, &last->done); - if (res == ERR_PTR(-ECHILD)) { - if (unlikely(unlazy_walk(nd))) - return ERR_PTR(-ECHILD); + if (res == ERR_PTR(-ECHILD) && try_to_unlazy(nd)) res = get(link->dentry, inode, &last->done); - } } else { res = get(link->dentry, inode, &last->done); } @@ -2195,7 +2190,7 @@ OK: } if (unlikely(!d_can_lookup(nd->path.dentry))) { if (nd->flags & LOOKUP_RCU) { - if (unlazy_walk(nd)) + if (!try_to_unlazy(nd)) return -ECHILD; } return -ENOTDIR; @@ -3129,7 +3124,6 @@ static const char *open_last_lookups(struct nameidata *nd, struct inode *inode; struct dentry *dentry; const char *res; - int error; nd->flags |= op->intent; @@ -3153,9 +3147,8 @@ static const char *open_last_lookups(struct nameidata *nd, } else { /* create side of things */ if (nd->flags & LOOKUP_RCU) { - error = unlazy_walk(nd); - if (unlikely(error)) - return ERR_PTR(error); + if (!try_to_unlazy(nd)) + return ERR_PTR(-ECHILD); } audit_inode(nd->name, dir, AUDIT_INODE_PARENT); /* trailing slashes? */ @@ -3164,9 +3157,7 @@ static const char *open_last_lookups(struct nameidata *nd, } if (open_flag & (O_CREAT | O_TRUNC | O_WRONLY | O_RDWR)) { - error = mnt_want_write(nd->path.mnt); - if (!error) - got_write = true; + got_write = !mnt_want_write(nd->path.mnt); /* * do _not_ fail yet - we might not need that or fail with * a different error; let lookup_open() decide; we'll be -- cgit From ae66db45fd309fd1c6d4e846dfc8414dfec7d6ad Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 4 Jan 2021 00:08:41 -0500 Subject: saner calling conventions for unlazy_child() same as for the previous commit - instead of 0/-ECHILD make it return true/false, rename to try_to_unlazy_child(). Signed-off-by: Al Viro --- fs/namei.c | 28 ++++++++++++++-------------- 1 file changed, 14 insertions(+), 14 deletions(-) diff --git a/fs/namei.c b/fs/namei.c index 21fd06753504..2ee219497460 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -705,19 +705,19 @@ out: } /** - * unlazy_child - try to switch to ref-walk mode. + * try_to_unlazy_next - try to switch to ref-walk mode. * @nd: nameidata pathwalk data - * @dentry: child of nd->path.dentry - * @seq: seq number to check dentry against - * Returns: 0 on success, -ECHILD on failure + * @dentry: next dentry to step into + * @seq: seq number to check @dentry against + * Returns: true on success, false on failure * - * unlazy_child attempts to legitimize the current nd->path, nd->root and dentry - * for ref-walk mode. @dentry must be a path found by a do_lookup call on - * @nd. Must be called from rcu-walk context. - * Nothing should touch nameidata between unlazy_child() failure and + * Similar to to try_to_unlazy(), but here we have the next dentry already + * picked by rcu-walk and want to legitimize that in addition to the current + * nd->path and nd->root for ref-walk mode. Must be called from rcu-walk context. + * Nothing should touch nameidata between try_to_unlazy_next() failure and * terminate_walk(). */ -static int unlazy_child(struct nameidata *nd, struct dentry *dentry, unsigned seq) +static bool try_to_unlazy_next(struct nameidata *nd, struct dentry *dentry, unsigned seq) { BUG_ON(!(nd->flags & LOOKUP_RCU)); @@ -747,7 +747,7 @@ static int unlazy_child(struct nameidata *nd, struct dentry *dentry, unsigned se if (unlikely(!legitimize_root(nd))) goto out_dput; rcu_read_unlock(); - return 0; + return true; out2: nd->path.mnt = NULL; @@ -755,11 +755,11 @@ out1: nd->path.dentry = NULL; out: rcu_read_unlock(); - return -ECHILD; + return false; out_dput: rcu_read_unlock(); dput(dentry); - return -ECHILD; + return false; } static inline int d_revalidate(struct dentry *dentry, unsigned int flags) @@ -1372,7 +1372,7 @@ static inline int handle_mounts(struct nameidata *nd, struct dentry *dentry, return -ENOENT; if (likely(__follow_mount_rcu(nd, path, inode, seqp))) return 0; - if (unlazy_child(nd, dentry, seq)) + if (!try_to_unlazy_next(nd, dentry, seq)) return -ECHILD; // *path might've been clobbered by __follow_mount_rcu() path->mnt = nd->path.mnt; @@ -1493,7 +1493,7 @@ static struct dentry *lookup_fast(struct nameidata *nd, status = d_revalidate(dentry, nd->flags); if (likely(status > 0)) return dentry; - if (unlazy_child(nd, dentry, seq)) + if (!try_to_unlazy_next(nd, dentry, seq)) return ERR_PTR(-ECHILD); if (status == -ECHILD) /* we'd been told to redo it in non-rcu mode */ -- cgit From 6c6ec2b0a3e0381d886d531bd1471dfdb1509237 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Thu, 17 Dec 2020 09:19:09 -0700 Subject: fs: add support for LOOKUP_CACHED io_uring always punts opens to async context, since there's no control over whether the lookup blocks or not. Add LOOKUP_CACHED to support just doing the fast RCU based lookups, which we know will not block. If we can do a cached path resolution of the filename, then we don't have to always punt lookups for a worker. During path resolution, we always do LOOKUP_RCU first. If that fails and we terminate LOOKUP_RCU, then fail a LOOKUP_CACHED attempt as well. Cc: Al Viro Signed-off-by: Jens Axboe Signed-off-by: Al Viro --- fs/namei.c | 9 +++++++++ include/linux/namei.h | 1 + 2 files changed, 10 insertions(+) diff --git a/fs/namei.c b/fs/namei.c index 2ee219497460..4cae88733a5c 100644 --- a/fs/namei.c +++ b/fs/namei.c @@ -686,6 +686,8 @@ static bool try_to_unlazy(struct nameidata *nd) BUG_ON(!(nd->flags & LOOKUP_RCU)); nd->flags &= ~LOOKUP_RCU; + if (nd->flags & LOOKUP_CACHED) + goto out1; if (unlikely(!legitimize_links(nd))) goto out1; if (unlikely(!legitimize_path(nd, &nd->path, nd->seq))) @@ -722,6 +724,8 @@ static bool try_to_unlazy_next(struct nameidata *nd, struct dentry *dentry, unsi BUG_ON(!(nd->flags & LOOKUP_RCU)); nd->flags &= ~LOOKUP_RCU; + if (nd->flags & LOOKUP_CACHED) + goto out2; if (unlikely(!legitimize_links(nd))) goto out2; if (unlikely(!legitimize_mnt(nd->path.mnt, nd->m_seq))) @@ -792,6 +796,7 @@ static int complete_walk(struct nameidata *nd) */ if (!(nd->flags & (LOOKUP_ROOT | LOOKUP_IS_SCOPED))) nd->root.mnt = NULL; + nd->flags &= ~LOOKUP_CACHED; if (!try_to_unlazy(nd)) return -ECHILD; } @@ -2204,6 +2209,10 @@ static const char *path_init(struct nameidata *nd, unsigned flags) int error; const char *s = nd->name->name; + /* LOOKUP_CACHED requires RCU, ask caller to retry */ + if ((flags & (LOOKUP_RCU | LOOKUP_CACHED)) == LOOKUP_CACHED) + return ERR_PTR(-EAGAIN); + if (!*s) flags &= ~LOOKUP_RCU; if (flags & LOOKUP_RCU) diff --git a/include/linux/namei.h b/include/linux/namei.h index a4bb992623c4..b9605b2b46e7 100644 --- a/include/linux/namei.h +++ b/include/linux/namei.h @@ -46,6 +46,7 @@ enum {LAST_NORM, LAST_ROOT, LAST_DOT, LAST_DOTDOT}; #define LOOKUP_NO_XDEV 0x040000 /* No mountpoint crossing. */ #define LOOKUP_BENEATH 0x080000 /* No escaping from starting point. */ #define LOOKUP_IN_ROOT 0x100000 /* Treat dirfd as fs root. */ +#define LOOKUP_CACHED 0x200000 /* Only do cached lookup */ /* LOOKUP_* flags which do scope-related checks based on the dirfd. */ #define LOOKUP_IS_SCOPED (LOOKUP_BENEATH | LOOKUP_IN_ROOT) -- cgit From 99668f618062816ca7ba639b007eb145b9d3d41e Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Thu, 17 Dec 2020 09:19:10 -0700 Subject: fs: expose LOOKUP_CACHED through openat2() RESOLVE_CACHED Now that we support non-blocking path resolution internally, expose it via openat2() in the struct open_how ->resolve flags. This allows applications using openat2() to limit path resolution to the extent that it is already cached. If the lookup cannot be satisfied in a non-blocking manner, openat2(2) will return -1/-EAGAIN. Cc: Al Viro Signed-off-by: Jens Axboe Signed-off-by: Al Viro --- fs/open.c | 6 ++++++ include/linux/fcntl.h | 2 +- include/uapi/linux/openat2.h | 4 ++++ 3 files changed, 11 insertions(+), 1 deletion(-) diff --git a/fs/open.c b/fs/open.c index 1e06e443a565..ca5444733acd 100644 --- a/fs/open.c +++ b/fs/open.c @@ -1091,6 +1091,12 @@ inline int build_open_flags(const struct open_how *how, struct open_flags *op) lookup_flags |= LOOKUP_BENEATH; if (how->resolve & RESOLVE_IN_ROOT) lookup_flags |= LOOKUP_IN_ROOT; + if (how->resolve & RESOLVE_CACHED) { + /* Don't bother even trying for create/truncate/tmpfile open */ + if (flags & (O_TRUNC | O_CREAT | O_TMPFILE)) + return -EAGAIN; + lookup_flags |= LOOKUP_CACHED; + } op->lookup_flags = lookup_flags; return 0; diff --git a/include/linux/fcntl.h b/include/linux/fcntl.h index 921e750843e6..766fcd973beb 100644 --- a/include/linux/fcntl.h +++ b/include/linux/fcntl.h @@ -19,7 +19,7 @@ /* List of all valid flags for the how->resolve argument: */ #define VALID_RESOLVE_FLAGS \ (RESOLVE_NO_XDEV | RESOLVE_NO_MAGICLINKS | RESOLVE_NO_SYMLINKS | \ - RESOLVE_BENEATH | RESOLVE_IN_ROOT) + RESOLVE_BENEATH | RESOLVE_IN_ROOT | RESOLVE_CACHED) /* List of all open_how "versions". */ #define OPEN_HOW_SIZE_VER0 24 /* sizeof first published struct */ diff --git a/include/uapi/linux/openat2.h b/include/uapi/linux/openat2.h index 58b1eb711360..a5feb7604948 100644 --- a/include/uapi/linux/openat2.h +++ b/include/uapi/linux/openat2.h @@ -35,5 +35,9 @@ struct open_how { #define RESOLVE_IN_ROOT 0x10 /* Make all jumps to "/" and ".." be scoped inside the dirfd (similar to chroot(2)). */ +#define RESOLVE_CACHED 0x20 /* Only complete if resolution can be + completed through cached lookup. May + return -EAGAIN if that's not + possible. */ #endif /* _UAPI_LINUX_OPENAT2_H */ -- cgit