| Message ID | 1342006085-5816-1-git-send-email-sprabhu@redhat.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
On Wed, Jul 11, 2012 at 6:28 AM, Sachin Prabhu <sprabhu@redhat.com> wrote: > A user reported a crash in cifs_demultiplex_thread() caused by an > incorrectly set mid_q_entry->callback() function. It appears that the > callback assignment made in cifs_call_async() was not flushed back to > memory suggesting that a memory barrier was required here. Changing the > code to make sure that the mid_q_entry structure was completely > initialised before it was added to the pending queue fixes the problem. > > Signed-off-by: Sachin Prabhu <sprabhu@redhat.com> > --- > fs/cifs/transport.c | 26 ++++++++++++++------------ > 1 file changed, 14 insertions(+), 12 deletions(-) > > diff --git a/fs/cifs/transport.c b/fs/cifs/transport.c > index 3097ee5..f25d4ea 100644 > --- a/fs/cifs/transport.c > +++ b/fs/cifs/transport.c > @@ -365,16 +365,14 @@ cifs_setup_async_request(struct TCP_Server_Info *server, struct kvec *iov, > if (mid == NULL) > return -ENOMEM; > > - /* put it on the pending_mid_q */ > - spin_lock(&GlobalMid_Lock); > - list_add_tail(&mid->qhead, &server->pending_mid_q); > - spin_unlock(&GlobalMid_Lock); > - > rc = cifs_sign_smb2(iov, nvec, server, &mid->sequence_number); > - if (rc) > - delete_mid(mid); > + if (rc) { > + DeleteMidQEntry(mid); > + return rc; > + } > + > *ret_mid = mid; > - return rc; > + return 0; > } > > /* > @@ -407,17 +405,21 @@ cifs_call_async(struct TCP_Server_Info *server, struct kvec *iov, > mid->callback_data = cbdata; > mid->mid_state = MID_REQUEST_SUBMITTED; > > + /* put it on the pending_mid_q */ > + spin_lock(&GlobalMid_Lock); > + list_add_tail(&mid->qhead, &server->pending_mid_q); > + spin_unlock(&GlobalMid_Lock); > + > + > cifs_in_send_inc(server); > rc = smb_sendv(server, iov, nvec); > cifs_in_send_dec(server); > cifs_save_when_sent(mid); > mutex_unlock(&server->srv_mutex); > > - if (rc) > - goto out_err; > + if (rc == 0) > + return 0; > > - return rc; > -out_err: > delete_mid(mid); > add_credits(server, 1); > wake_up(&server->request_q); > -- > 1.7.10.4 > Looks correct. Reviewed-by: Shirish Pargaonkar <shirishpargaonkar@gmail.com> -- To unsubscribe from this list: send the line "unsubscribe linux-cifs" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On Wed, 11 Jul 2012 12:28:05 +0100 Sachin Prabhu <sprabhu@redhat.com> wrote: > A user reported a crash in cifs_demultiplex_thread() caused by an > incorrectly set mid_q_entry->callback() function. It appears that the > callback assignment made in cifs_call_async() was not flushed back to > memory suggesting that a memory barrier was required here. Changing the > code to make sure that the mid_q_entry structure was completely > initialised before it was added to the pending queue fixes the problem. > > Signed-off-by: Sachin Prabhu <sprabhu@redhat.com> > --- > fs/cifs/transport.c | 26 ++++++++++++++------------ > 1 file changed, 14 insertions(+), 12 deletions(-) > > diff --git a/fs/cifs/transport.c b/fs/cifs/transport.c > index 3097ee5..f25d4ea 100644 > --- a/fs/cifs/transport.c > +++ b/fs/cifs/transport.c > @@ -365,16 +365,14 @@ cifs_setup_async_request(struct TCP_Server_Info *server, struct kvec *iov, > if (mid == NULL) > return -ENOMEM; > > - /* put it on the pending_mid_q */ > - spin_lock(&GlobalMid_Lock); > - list_add_tail(&mid->qhead, &server->pending_mid_q); > - spin_unlock(&GlobalMid_Lock); > - > rc = cifs_sign_smb2(iov, nvec, server, &mid->sequence_number); > - if (rc) > - delete_mid(mid); > + if (rc) { > + DeleteMidQEntry(mid); > + return rc; > + } > + > *ret_mid = mid; > - return rc; > + return 0; > } > > /* > @@ -407,17 +405,21 @@ cifs_call_async(struct TCP_Server_Info *server, struct kvec *iov, > mid->callback_data = cbdata; > mid->mid_state = MID_REQUEST_SUBMITTED; > > + /* put it on the pending_mid_q */ > + spin_lock(&GlobalMid_Lock); > + list_add_tail(&mid->qhead, &server->pending_mid_q); > + spin_unlock(&GlobalMid_Lock); > + > + > cifs_in_send_inc(server); > rc = smb_sendv(server, iov, nvec); > cifs_in_send_dec(server); > cifs_save_when_sent(mid); > mutex_unlock(&server->srv_mutex); > > - if (rc) > - goto out_err; > + if (rc == 0) > + return 0; > > - return rc; > -out_err: > delete_mid(mid); > add_credits(server, 1); > wake_up(&server->request_q); This version looks good to me. I think this ought to go to stable too. Reviewed-by: Jeff Layton <jlayton@redhat.com> -- To unsubscribe from this list: send the line "unsubscribe linux-cifs" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
diff --git a/fs/cifs/transport.c b/fs/cifs/transport.c index 3097ee5..f25d4ea 100644 --- a/fs/cifs/transport.c +++ b/fs/cifs/transport.c @@ -365,16 +365,14 @@ cifs_setup_async_request(struct TCP_Server_Info *server, struct kvec *iov, if (mid == NULL) return -ENOMEM; - /* put it on the pending_mid_q */ - spin_lock(&GlobalMid_Lock); - list_add_tail(&mid->qhead, &server->pending_mid_q); - spin_unlock(&GlobalMid_Lock); - rc = cifs_sign_smb2(iov, nvec, server, &mid->sequence_number); - if (rc) - delete_mid(mid); + if (rc) { + DeleteMidQEntry(mid); + return rc; + } + *ret_mid = mid; - return rc; + return 0; } /* @@ -407,17 +405,21 @@ cifs_call_async(struct TCP_Server_Info *server, struct kvec *iov, mid->callback_data = cbdata; mid->mid_state = MID_REQUEST_SUBMITTED; + /* put it on the pending_mid_q */ + spin_lock(&GlobalMid_Lock); + list_add_tail(&mid->qhead, &server->pending_mid_q); + spin_unlock(&GlobalMid_Lock); + + cifs_in_send_inc(server); rc = smb_sendv(server, iov, nvec); cifs_in_send_dec(server); cifs_save_when_sent(mid); mutex_unlock(&server->srv_mutex); - if (rc) - goto out_err; + if (rc == 0) + return 0; - return rc; -out_err: delete_mid(mid); add_credits(server, 1); wake_up(&server->request_q);
A user reported a crash in cifs_demultiplex_thread() caused by an incorrectly set mid_q_entry->callback() function. It appears that the callback assignment made in cifs_call_async() was not flushed back to memory suggesting that a memory barrier was required here. Changing the code to make sure that the mid_q_entry structure was completely initialised before it was added to the pending queue fixes the problem. Signed-off-by: Sachin Prabhu <sprabhu@redhat.com> --- fs/cifs/transport.c | 26 ++++++++++++++------------ 1 file changed, 14 insertions(+), 12 deletions(-)