mbox series

[v4,00/17] libiscsi and qedi TMF fixes

Message ID 20210424220603.123703-1-michael.christie@oracle.com
Headers show
Series libiscsi and qedi TMF fixes | expand

Message

Mike Christie April 24, 2021, 10:05 p.m. UTC 
The following patches were made over Linus's tree. They apply to any
of Martin's branches that has this fix: 73603e995a37 ("scsi: iscsi: Fix
iSCSI cls conn state") that has conflicts with a patch in this set.

The patches fix TMF bugs in the qedi driver, libiscsi EH issues that are
common to all offload drivers like qedi and some fixes for cases where
userspace is not doing an unbind target nl cmd and we are doing TMFs
during session termination.


V4:
- Rm unused rtid variable in qedi
- Added a default case in TMF switch just in case people do not like
not having it.
V3:
- Fix u16 initialization and test.
- Fix bool return value use.
- Added patches for cases where EH is running then userspace terminates
the connection without removing the target first.
- Made patch that stops IO during ep_disconnect more driver friendly
by handling if the ep is bound or not.

V2:
- Dropped patch that reverted the in-kernel conn failure handling. I
posted a full patchset to fix that separately.
- Modfied the tmf vs cmd queueing patch. The RFC patch only supported
qedi and offload drivers. iscsi_tcp/cxgbgi do not need it.
- Added patch for another issue found where cmds can still be queued to the
driver while it does ep_disconnect.

Comments

Mike Christie May 5, 2021, 3:02 a.m. UTC | #1 
On 4/24/21 5:05 PM, Mike Christie wrote:
> The following patches were made over Linus's tree. They apply to any

> of Martin's branches that has this fix: 73603e995a37 ("scsi: iscsi: Fix

> iSCSI cls conn state") that has conflicts with a patch in this set.

> 

> The patches fix TMF bugs in the qedi driver, libiscsi EH issues that are

> common to all offload drivers like qedi and some fixes for cases where

> userspace is not doing an unbind target nl cmd and we are doing TMFs

> during session termination.

> 



Hi Martin,

Please don't merge this patchset. I want to make a change that will make
my other iscsi patchsets easier to merge, and fix a patch so we don't
get multiple error messages for the same problem.

When Lee is done reviewing "[PATCH v3 0/6] iscsi: Fix in kernel conn
failure handling" I'll roll everything up into one patchset.
Martin K. Petersen May 5, 2021, 3:18 a.m. UTC | #2 
> Please don't merge this patchset. I want to make a change that will

> make my other iscsi patchsets easier to merge, and fix a patch so we

> don't get multiple error messages for the same problem.

>

> When Lee is done reviewing "[PATCH v3 0/6] iscsi: Fix in kernel conn

> failure handling" I'll roll everything up into one patchset.


OK. It's currently sitting in staging but I'll just drop it again. I
would have had to rebase it regardless due to the DOWN vs. FAILED
conflict.

-- 
Martin K. Petersen	Oracle Linux Engineering