| Message ID | 1502577017-1236-2-git-send-email-mchristi@redhat.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
On 08/12/2017 05:30 PM, Mike Christie wrote: > If userspace is disabling a tpg via the tpg enable configfs file > and the initiator or target decides to drop the iscsi session, then > we can end up where only one thread waiting on session_wait_comp will > be woken up. This can then lead to the app doing the disable to hang > if it was the one that did not get notified. > > Signed-off-by: Mike Christie <mchristi@redhat.com> > --- > drivers/target/iscsi/iscsi_target.c | 4 ++-- > 1 file changed, 2 insertions(+), 2 deletions(-) > > diff --git a/drivers/target/iscsi/iscsi_target.c b/drivers/target/iscsi/iscsi_target.c > index 5001261..e1e0e25 100644 > --- a/drivers/target/iscsi/iscsi_target.c > +++ b/drivers/target/iscsi/iscsi_target.c > @@ -4318,7 +4318,7 @@ int iscsit_close_connection( > spin_unlock_bh(&sess->conn_lock); > > if (atomic_read(&sess->sleep_on_sess_wait_comp)) > - complete(&sess->session_wait_comp); > + complete_all(&sess->session_wait_comp); > > return 0; > } else { > @@ -4332,7 +4332,7 @@ int iscsit_close_connection( > spin_unlock_bh(&sess->conn_lock); > > if (atomic_read(&sess->sleep_on_sess_wait_comp)) > - complete(&sess->session_wait_comp); > + complete_all(&sess->session_wait_comp); > > return 0; > } > Nick, ignore this patch for now. The bug might only be in a distro kernel and is not needed upstream. -- To unsubscribe from this list: send the line "unsubscribe target-devel" 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/drivers/target/iscsi/iscsi_target.c b/drivers/target/iscsi/iscsi_target.c index 5001261..e1e0e25 100644 --- a/drivers/target/iscsi/iscsi_target.c +++ b/drivers/target/iscsi/iscsi_target.c @@ -4318,7 +4318,7 @@ int iscsit_close_connection( spin_unlock_bh(&sess->conn_lock); if (atomic_read(&sess->sleep_on_sess_wait_comp)) - complete(&sess->session_wait_comp); + complete_all(&sess->session_wait_comp); return 0; } else { @@ -4332,7 +4332,7 @@ int iscsit_close_connection( spin_unlock_bh(&sess->conn_lock); if (atomic_read(&sess->sleep_on_sess_wait_comp)) - complete(&sess->session_wait_comp); + complete_all(&sess->session_wait_comp); return 0; }
If userspace is disabling a tpg via the tpg enable configfs file and the initiator or target decides to drop the iscsi session, then we can end up where only one thread waiting on session_wait_comp will be woken up. This can then lead to the app doing the disable to hang if it was the one that did not get notified. Signed-off-by: Mike Christie <mchristi@redhat.com> --- drivers/target/iscsi/iscsi_target.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)