| Message ID | 20230609174659.60327-1-sprasad@microsoft.com |
|---|---|
| State | New |
| Headers | show |
| Series | [1/6] cifs: fix status checks in cifs_tree_connect | expand |
tentatively merged into cifs-2.6.git for-next On Fri, Jun 9, 2023 at 12:47 PM Shyam Prasad N <nspmangalore@gmail.com> wrote: > > The ordering of status checks at the beginning of > cifs_tree_connect is wrong. As a result, a tcon > which is good may stay marked as needing reconnect > infinitely. > > Fixes: 2f0e4f034220 ("cifs: check only tcon status on tcon related functions") > Signed-off-by: Shyam Prasad N <sprasad@microsoft.com> > --- > fs/smb/client/connect.c | 9 +++++---- > fs/smb/client/dfs.c | 9 +++++---- > 2 files changed, 10 insertions(+), 8 deletions(-) > > diff --git a/fs/smb/client/connect.c b/fs/smb/client/connect.c > index 8e9a672320ab..1250d156619b 100644 > --- a/fs/smb/client/connect.c > +++ b/fs/smb/client/connect.c > @@ -4086,16 +4086,17 @@ int cifs_tree_connect(const unsigned int xid, struct cifs_tcon *tcon, const stru > > /* only send once per connect */ > spin_lock(&tcon->tc_lock); > + if (tcon->status == TID_GOOD) { > + spin_unlock(&tcon->tc_lock); > + return 0; > + } > + > if (tcon->status != TID_NEW && > tcon->status != TID_NEED_TCON) { > spin_unlock(&tcon->tc_lock); > return -EHOSTDOWN; > } > > - if (tcon->status == TID_GOOD) { > - spin_unlock(&tcon->tc_lock); > - return 0; > - } > tcon->status = TID_IN_TCON; > spin_unlock(&tcon->tc_lock); > > diff --git a/fs/smb/client/dfs.c b/fs/smb/client/dfs.c > index 2f93bf8c3325..2390b2fedd6a 100644 > --- a/fs/smb/client/dfs.c > +++ b/fs/smb/client/dfs.c > @@ -575,16 +575,17 @@ int cifs_tree_connect(const unsigned int xid, struct cifs_tcon *tcon, const stru > > /* only send once per connect */ > spin_lock(&tcon->tc_lock); > + if (tcon->status == TID_GOOD) { > + spin_unlock(&tcon->tc_lock); > + return 0; > + } > + > if (tcon->status != TID_NEW && > tcon->status != TID_NEED_TCON) { > spin_unlock(&tcon->tc_lock); > return -EHOSTDOWN; > } > > - if (tcon->status == TID_GOOD) { > - spin_unlock(&tcon->tc_lock); > - return 0; > - } > tcon->status = TID_IN_TCON; > spin_unlock(&tcon->tc_lock); > > -- > 2.34.1 >
diff --git a/fs/smb/client/connect.c b/fs/smb/client/connect.c index 8e9a672320ab..1250d156619b 100644 --- a/fs/smb/client/connect.c +++ b/fs/smb/client/connect.c @@ -4086,16 +4086,17 @@ int cifs_tree_connect(const unsigned int xid, struct cifs_tcon *tcon, const stru /* only send once per connect */ spin_lock(&tcon->tc_lock); + if (tcon->status == TID_GOOD) { + spin_unlock(&tcon->tc_lock); + return 0; + } + if (tcon->status != TID_NEW && tcon->status != TID_NEED_TCON) { spin_unlock(&tcon->tc_lock); return -EHOSTDOWN; } - if (tcon->status == TID_GOOD) { - spin_unlock(&tcon->tc_lock); - return 0; - } tcon->status = TID_IN_TCON; spin_unlock(&tcon->tc_lock); diff --git a/fs/smb/client/dfs.c b/fs/smb/client/dfs.c index 2f93bf8c3325..2390b2fedd6a 100644 --- a/fs/smb/client/dfs.c +++ b/fs/smb/client/dfs.c @@ -575,16 +575,17 @@ int cifs_tree_connect(const unsigned int xid, struct cifs_tcon *tcon, const stru /* only send once per connect */ spin_lock(&tcon->tc_lock); + if (tcon->status == TID_GOOD) { + spin_unlock(&tcon->tc_lock); + return 0; + } + if (tcon->status != TID_NEW && tcon->status != TID_NEED_TCON) { spin_unlock(&tcon->tc_lock); return -EHOSTDOWN; } - if (tcon->status == TID_GOOD) { - spin_unlock(&tcon->tc_lock); - return 0; - } tcon->status = TID_IN_TCON; spin_unlock(&tcon->tc_lock);
The ordering of status checks at the beginning of cifs_tree_connect is wrong. As a result, a tcon which is good may stay marked as needing reconnect infinitely. Fixes: 2f0e4f034220 ("cifs: check only tcon status on tcon related functions") Signed-off-by: Shyam Prasad N <sprasad@microsoft.com> --- fs/smb/client/connect.c | 9 +++++---- fs/smb/client/dfs.c | 9 +++++---- 2 files changed, 10 insertions(+), 8 deletions(-)