From: David Howells <dhowells@redhat.com> To: linux-cachefs@redhat.com Cc: Gao Xiang <hsiangkao@linux.alibaba.com>, Jeffle Xu <jefflexu@linux.alibaba.com>, dhowells@redhat.com, Jeff Layton <jlayton@kernel.org>, Jeffle Xu <jefflexu@linux.alibaba.com>, Yue Hu <huyue2@coolpad.com>, linux-fsdevel@vger.kernel.org, linux-erofs@lists.ozlabs.org, linux-kernel@vger.kernel.org Subject: [PATCH 1/8] cachefiles: unmark inode in use in error path Date: Sat, 09 Apr 2022 00:06:10 +0100 [thread overview] Message-ID: <164945917002.773423.16291667819231402497.stgit@warthog.procyon.org.uk> (raw) In-Reply-To: <164945915630.773423.14655306154231712324.stgit@warthog.procyon.org.uk> From: Jeffle Xu <jefflexu@linux.alibaba.com> Unmark inode in use if error encountered. If the in-use flag leakage occurs in cachefiles_open_file(), Cachefiles will complain "Inode already in use" when later another cookie with the same index key is looked up. If the in-use flag leakage occurs in cachefiles_create_tmpfile(), though the "Inode already in use" warning won't be triggered, fix the leakage anyway. Reported-by: Gao Xiang <hsiangkao@linux.alibaba.com> Fixes: 1f08c925e7a3 ("cachefiles: Implement backing file wrangling") Signed-off-by: Jeffle Xu <jefflexu@linux.alibaba.com> Signed-off-by: David Howells <dhowells@redhat.com> cc: linux-cachefs@redhat.com Link: https://listman.redhat.com/archives/linux-cachefs/2022-March/006615.html # v1 Link: https://listman.redhat.com/archives/linux-cachefs/2022-March/006618.html # v2 --- fs/cachefiles/namei.c | 33 ++++++++++++++++++++++++--------- 1 file changed, 24 insertions(+), 9 deletions(-) diff --git a/fs/cachefiles/namei.c b/fs/cachefiles/namei.c index f256c8aff7bb..ca9f3e4ec4b3 100644 --- a/fs/cachefiles/namei.c +++ b/fs/cachefiles/namei.c @@ -57,6 +57,16 @@ static void __cachefiles_unmark_inode_in_use(struct cachefiles_object *object, trace_cachefiles_mark_inactive(object, inode); } +static void cachefiles_do_unmark_inode_in_use(struct cachefiles_object *object, + struct dentry *dentry) +{ + struct inode *inode = d_backing_inode(dentry); + + inode_lock(inode); + __cachefiles_unmark_inode_in_use(object, dentry); + inode_unlock(inode); +} + /* * Unmark a backing inode and tell cachefilesd that there's something that can * be culled. @@ -68,9 +78,7 @@ void cachefiles_unmark_inode_in_use(struct cachefiles_object *object, struct inode *inode = file_inode(file); if (inode) { - inode_lock(inode); - __cachefiles_unmark_inode_in_use(object, file->f_path.dentry); - inode_unlock(inode); + cachefiles_do_unmark_inode_in_use(object, file->f_path.dentry); if (!test_bit(CACHEFILES_OBJECT_USING_TMPFILE, &object->flags)) { atomic_long_add(inode->i_blocks, &cache->b_released); @@ -484,7 +492,7 @@ struct file *cachefiles_create_tmpfile(struct cachefiles_object *object) object, d_backing_inode(path.dentry), ret, cachefiles_trace_trunc_error); file = ERR_PTR(ret); - goto out_dput; + goto out_unuse; } } @@ -494,15 +502,20 @@ struct file *cachefiles_create_tmpfile(struct cachefiles_object *object) trace_cachefiles_vfs_error(object, d_backing_inode(path.dentry), PTR_ERR(file), cachefiles_trace_open_error); - goto out_dput; + goto out_unuse; } if (unlikely(!file->f_op->read_iter) || unlikely(!file->f_op->write_iter)) { fput(file); pr_notice("Cache does not support read_iter and write_iter\n"); file = ERR_PTR(-EINVAL); + goto out_unuse; } + goto out_dput; + +out_unuse: + cachefiles_do_unmark_inode_in_use(object, path.dentry); out_dput: dput(path.dentry); out: @@ -590,14 +603,16 @@ static bool cachefiles_open_file(struct cachefiles_object *object, check_failed: fscache_cookie_lookup_negative(object->cookie); cachefiles_unmark_inode_in_use(object, file); - if (ret == -ESTALE) { - fput(file); - dput(dentry); + fput(file); + dput(dentry); + if (ret == -ESTALE) return cachefiles_create_file(object); - } + return false; + error_fput: fput(file); error: + cachefiles_do_unmark_inode_in_use(object, dentry); dput(dentry); return false; }
WARNING: multiple messages have this Message-ID (diff)
From: David Howells <dhowells@redhat.com> To: linux-cachefs@redhat.com Cc: Jeff Layton <jlayton@kernel.org>, linux-kernel@vger.kernel.org, dhowells@redhat.com, linux-fsdevel@vger.kernel.org, Yue Hu <huyue2@coolpad.com>, Gao Xiang <hsiangkao@linux.alibaba.com>, linux-erofs@lists.ozlabs.org Subject: [PATCH 1/8] cachefiles: unmark inode in use in error path Date: Sat, 09 Apr 2022 00:06:10 +0100 [thread overview] Message-ID: <164945917002.773423.16291667819231402497.stgit@warthog.procyon.org.uk> (raw) In-Reply-To: <164945915630.773423.14655306154231712324.stgit@warthog.procyon.org.uk> From: Jeffle Xu <jefflexu@linux.alibaba.com> Unmark inode in use if error encountered. If the in-use flag leakage occurs in cachefiles_open_file(), Cachefiles will complain "Inode already in use" when later another cookie with the same index key is looked up. If the in-use flag leakage occurs in cachefiles_create_tmpfile(), though the "Inode already in use" warning won't be triggered, fix the leakage anyway. Reported-by: Gao Xiang <hsiangkao@linux.alibaba.com> Fixes: 1f08c925e7a3 ("cachefiles: Implement backing file wrangling") Signed-off-by: Jeffle Xu <jefflexu@linux.alibaba.com> Signed-off-by: David Howells <dhowells@redhat.com> cc: linux-cachefs@redhat.com Link: https://listman.redhat.com/archives/linux-cachefs/2022-March/006615.html # v1 Link: https://listman.redhat.com/archives/linux-cachefs/2022-March/006618.html # v2 --- fs/cachefiles/namei.c | 33 ++++++++++++++++++++++++--------- 1 file changed, 24 insertions(+), 9 deletions(-) diff --git a/fs/cachefiles/namei.c b/fs/cachefiles/namei.c index f256c8aff7bb..ca9f3e4ec4b3 100644 --- a/fs/cachefiles/namei.c +++ b/fs/cachefiles/namei.c @@ -57,6 +57,16 @@ static void __cachefiles_unmark_inode_in_use(struct cachefiles_object *object, trace_cachefiles_mark_inactive(object, inode); } +static void cachefiles_do_unmark_inode_in_use(struct cachefiles_object *object, + struct dentry *dentry) +{ + struct inode *inode = d_backing_inode(dentry); + + inode_lock(inode); + __cachefiles_unmark_inode_in_use(object, dentry); + inode_unlock(inode); +} + /* * Unmark a backing inode and tell cachefilesd that there's something that can * be culled. @@ -68,9 +78,7 @@ void cachefiles_unmark_inode_in_use(struct cachefiles_object *object, struct inode *inode = file_inode(file); if (inode) { - inode_lock(inode); - __cachefiles_unmark_inode_in_use(object, file->f_path.dentry); - inode_unlock(inode); + cachefiles_do_unmark_inode_in_use(object, file->f_path.dentry); if (!test_bit(CACHEFILES_OBJECT_USING_TMPFILE, &object->flags)) { atomic_long_add(inode->i_blocks, &cache->b_released); @@ -484,7 +492,7 @@ struct file *cachefiles_create_tmpfile(struct cachefiles_object *object) object, d_backing_inode(path.dentry), ret, cachefiles_trace_trunc_error); file = ERR_PTR(ret); - goto out_dput; + goto out_unuse; } } @@ -494,15 +502,20 @@ struct file *cachefiles_create_tmpfile(struct cachefiles_object *object) trace_cachefiles_vfs_error(object, d_backing_inode(path.dentry), PTR_ERR(file), cachefiles_trace_open_error); - goto out_dput; + goto out_unuse; } if (unlikely(!file->f_op->read_iter) || unlikely(!file->f_op->write_iter)) { fput(file); pr_notice("Cache does not support read_iter and write_iter\n"); file = ERR_PTR(-EINVAL); + goto out_unuse; } + goto out_dput; + +out_unuse: + cachefiles_do_unmark_inode_in_use(object, path.dentry); out_dput: dput(path.dentry); out: @@ -590,14 +603,16 @@ static bool cachefiles_open_file(struct cachefiles_object *object, check_failed: fscache_cookie_lookup_negative(object->cookie); cachefiles_unmark_inode_in_use(object, file); - if (ret == -ESTALE) { - fput(file); - dput(dentry); + fput(file); + dput(dentry); + if (ret == -ESTALE) return cachefiles_create_file(object); - } + return false; + error_fput: fput(file); error: + cachefiles_do_unmark_inode_in_use(object, dentry); dput(dentry); return false; }
next prev parent reply other threads:[~2022-04-08 23:06 UTC|newest] Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top 2022-04-08 23:05 [RFC][PATCH 0/8] fscache, cachefiles: Fixes David Howells 2022-04-08 23:05 ` David Howells 2022-04-08 23:06 ` David Howells [this message] 2022-04-08 23:06 ` [PATCH 1/8] cachefiles: unmark inode in use in error path David Howells 2022-04-08 23:06 ` [PATCH 2/8] cachefiles: Fix KASAN slab-out-of-bounds in cachefiles_set_volume_xattr David Howells 2022-04-08 23:06 ` David Howells 2022-04-08 23:06 ` [PATCH 3/8] docs: filesystems: caching/backend-api.rst: correct two relinquish APIs use David Howells 2022-04-08 23:06 ` David Howells 2022-04-08 23:06 ` [PATCH 4/8] docs: filesystems: caching/backend-api.rst: fix an object withdrawn API David Howells 2022-04-08 23:06 ` David Howells 2022-04-08 23:06 ` [PATCH 5/8] fscache: Remove the cookie parameter from fscache_clear_page_bits() David Howells 2022-04-08 23:06 ` David Howells 2022-04-08 23:06 ` [PATCH 6/8] fscache: Move fscache_cookies_seq_ops specific code under CONFIG_PROC_FS David Howells 2022-04-08 23:06 ` David Howells 2022-04-08 23:07 ` [PATCH 7/8] fscache: Use wrapper fscache_set_cache_state() directly when relinquishing David Howells 2022-04-08 23:07 ` David Howells 2022-04-08 23:07 ` [PATCH 8/8] fscache: remove FSCACHE_OLD_API Kconfig option David Howells 2022-04-08 23:07 ` David Howells
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to=164945917002.773423.16291667819231402497.stgit@warthog.procyon.org.uk \ --to=dhowells@redhat.com \ --cc=hsiangkao@linux.alibaba.com \ --cc=huyue2@coolpad.com \ --cc=jefflexu@linux.alibaba.com \ --cc=jlayton@kernel.org \ --cc=linux-cachefs@redhat.com \ --cc=linux-erofs@lists.ozlabs.org \ --cc=linux-fsdevel@vger.kernel.org \ --cc=linux-kernel@vger.kernel.org \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.