Messages in this thread | | | Date | Wed, 14 Apr 2021 08:14:40 -0400 | From | Sasha Levin <> | Subject | Re: [PATCH AUTOSEL 5.10 07/22] scsi: iscsi: Fix race condition between login and sync thread |
| |
On Tue, Apr 06, 2021 at 12:24:32PM -0500, Mike Christie wrote: >On 4/5/21 11:04 AM, Sasha Levin wrote: >> From: Gulam Mohamed <gulam.mohamed@oracle.com> >> >> [ Upstream commit 9e67600ed6b8565da4b85698ec659b5879a6c1c6 ] >> >> A kernel panic was observed due to a timing issue between the sync thread >> and the initiator processing a login response from the target. The session >> reopen can be invoked both from the session sync thread when iscsid >> restarts and from iscsid through the error handler. Before the initiator >> receives the response to a login, another reopen request can be sent from >> the error handler/sync session. When the initial login response is >> subsequently processed, the connection has been closed and the socket has >> been released. >> >> To fix this a new connection state, ISCSI_CONN_BOUND, is added: >> >> - Set the connection state value to ISCSI_CONN_DOWN upon >> iscsi_if_ep_disconnect() and iscsi_if_stop_conn() >> >> - Set the connection state to the newly created value ISCSI_CONN_BOUND >> after bind connection (transport->bind_conn()) >> >> - In iscsi_set_param(), return -ENOTCONN if the connection state is not >> either ISCSI_CONN_BOUND or ISCSI_CONN_UP >> >> Link: https://urldefense.com/v3/__https://lore.kernel.org/r/20210325093248.284678-1-gulam.mohamed@oracle.com__;!!GqivPVa7Brio!Jiqrc6pu3EgrquzpG-KpNQkNebwKUgctkE0MN1MloQ2y5Y4OVOkKN0yCr2_W_CX2oRet$ >> Reviewed-by: Mike Christie <michael.christie@oracle.com> > > >There was a mistake in my review of this patch. It will also require >this "[PATCH 1/1] scsi: iscsi: fix iscsi cls conn state": > >https://lore.kernel.org/linux-scsi/20210406171746.5016-1-michael.christie@oracle.com/T/#u
As the fix isn't upstream yet, I'll drop 9e67600ed6b for now and re-queue it for the next round. Thanks!
-- Thanks, Sasha
| |